PlantGeneBench-100 题库修订审阅

Rationale

intent=`gene_family_identification`; entities: species=rice, family=WRKY; target_stages=[identify]; execution_plan=[identify]; required_inputs=[protein_fasta,hmm_profile]; can_execute=true if files exist.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {
      "geneFamily": "WRKY",
      "species": "rice"
    },
    "species": "rice",
    "stages": [
      "identify"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {
        "geneFamily": "WRKY",
        "species": "rice"
      },
      "species": "rice",
      "stages": [
        "identify"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-001",
  "user_query": "“我想鉴定水稻 WRKY 基因家族成员。”上下文：有 `rice_protein.fa` 和 `WRKY.hmm`。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "protein_fasta",
      "hmm_profile"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": true,
  "entities": {
    "gene_family": "WRKY",
    "species": "rice"
  },
  "evidence": [],
  "execution_plan": [
    "identify"
  ],
  "expected_execution_plan": [
    "identify"
  ],
  "expected_stages": [
    "identify"
  ],
  "forbidden_stages": [
    "expression",
    "rnaseq_align",
    "rnaseq_de",
    "rnaseq_enrich",
    "rnaseq_qc",
    "rnaseq_quant"
  ],
  "intent": "gene_family_identification",
  "must_include": [],
  "must_not_do": [
    "不要安排 expression、motif 或 RNA-seq",
    "不要声称已完成鉴定。"
  ],
  "parameters": {
    "geneFamily": "WRKY",
    "species": "rice"
  },
  "rationale": "intent=`gene_family_identification`; entities: species=rice, family=WRKY; target_stages=[identify]; execution_plan=[identify]; required_inputs=[protein_fasta,hmm_profile]; can_execute=true if files exist.",
  "required_dependencies": {},
  "required_params": {
    "geneFamily": "WRKY",
    "species": "rice"
  },
  "support_status": null,
  "target_stages": [
    "identify"
  ]
}

Rationale

intent=`gene_family_phylogeny`; species=Arabidopsis; family=SPL; target_stages=[phylogeny]; execution_plan=[phylogeny]; selected_input=member_protein_fasta.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "SPL",
    "params": {
      "geneFamily": "SPL",
      "species": "Arabidopsis"
    },
    "species": "Arabidopsis",
    "stages": [
      "phylogeny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "SPL",
      "params": {
        "geneFamily": "SPL",
        "species": "Arabidopsis"
      },
      "species": "Arabidopsis",
      "stages": [
        "phylogeny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-002",
  "user_query": "“帮我做拟南芥 SPL 的系统发育树。”上下文：已有 `At.SPL.final_members.pep.fa`。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "SPL",
    "species": "Arabidopsis"
  },
  "evidence": [],
  "execution_plan": [
    "phylogeny"
  ],
  "expected_execution_plan": [
    "phylogeny"
  ],
  "expected_stages": [
    "phylogeny"
  ],
  "forbidden_stages": [
    "identify"
  ],
  "intent": "gene_family_phylogeny",
  "must_include": [],
  "must_not_do": [
    "不要重新跑 identify 作为必需步骤",
    "不要把 phylogeny 误判为 tree_plot-only。"
  ],
  "parameters": {
    "geneFamily": "SPL",
    "species": "Arabidopsis"
  },
  "rationale": "intent=`gene_family_phylogeny`; species=Arabidopsis; family=SPL; target_stages=[phylogeny]; execution_plan=[phylogeny]; selected_input=member_protein_fasta.",
  "required_dependencies": {},
  "required_params": {
    "geneFamily": "SPL",
    "species": "Arabidopsis"
  },
  "support_status": null,
  "target_stages": [
    "phylogeny"
  ]
}

Rationale

intent=`gene_family_motif_analysis`; family=NAC; target_stages=[motif]; execution_plan=[motif]; selected_input=member_protein_fasta; default_tool=MEME-compatible motif stage.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "NAC",
    "params": {
      "geneFamily": "NAC"
    },
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "NAC",
      "params": {
        "geneFamily": "NAC"
      },
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-003",
  "user_query": "“用这些 NAC 蛋白序列找保守 motif。”上下文：`NAC.members.pep.fa`。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "NAC"
  },
  "evidence": [],
  "execution_plan": [
    "motif"
  ],
  "expected_execution_plan": [
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "gene_family_motif_analysis",
  "must_include": [],
  "must_not_do": [
    "不要用全蛋白组代替成员蛋白",
    "不要要求 GFF 作为 motif 必需输入。"
  ],
  "parameters": {
    "geneFamily": "NAC"
  },
  "rationale": "intent=`gene_family_motif_analysis`; family=NAC; target_stages=[motif]; execution_plan=[motif]; selected_input=member_protein_fasta; default_tool=MEME-compatible motif stage.",
  "required_dependencies": {},
  "required_params": {
    "geneFamily": "NAC"
  },
  "support_status": null,
  "target_stages": [
    "motif"
  ]
}

Rationale

intent=`gene_structure_analysis`; family=WRKY; target_stages=[structure]; execution_plan=[structure]; required_inputs=[member_ids,gff3].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {
      "geneFamily": "WRKY"
    },
    "species": null,
    "stages": [
      "structure"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {
        "geneFamily": "WRKY"
      },
      "species": null,
      "stages": [
        "structure"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-004",
  "user_query": "“我想看 WRKY 基因结构，已有成员 ID 和 GFF3。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_ids",
      "gff3"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "structure"
  ],
  "expected_execution_plan": [
    "structure"
  ],
  "expected_stages": [
    "structure"
  ],
  "forbidden_stages": [
    "motif"
  ],
  "intent": "gene_structure_analysis",
  "must_include": [],
  "must_not_do": [
    "不要转到 motif/domain",
    "不要说只有 CDS 才能画结构。"
  ],
  "parameters": {
    "geneFamily": "WRKY"
  },
  "rationale": "intent=`gene_structure_analysis`; family=WRKY; target_stages=[structure]; execution_plan=[structure]; required_inputs=[member_ids,gff3].",
  "required_dependencies": {},
  "required_params": {
    "geneFamily": "WRKY"
  },
  "support_status": null,
  "target_stages": [
    "structure"
  ]
}

Rationale

intent=`promoter_cis_element_analysis`; target_stages=[promoter]; parameters.promoter_length=2000; required_inputs=[genome_fasta,gff3,member_ids].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {
      "promoter_length": 2000
    },
    "species": null,
    "stages": [
      "promoter"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {
        "promoter_length": 2000
      },
      "species": null,
      "stages": [
        "promoter"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-005",
  "user_query": "“做启动子顺式元件分析，上游 2000 bp。”上下文：有 genome FASTA、GFF3 和成员 ID。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "genome_fasta",
      "gff3",
      "member_ids"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "promoter"
  ],
  "expected_execution_plan": [
    "promoter"
  ],
  "expected_stages": [
    "promoter"
  ],
  "forbidden_stages": [],
  "intent": "promoter_cis_element_analysis",
  "must_include": [],
  "must_not_do": [
    "不要默认 1500 bp",
    "不要忽略上游长度。"
  ],
  "parameters": {
    "promoter_length": 2000
  },
  "rationale": "intent=`promoter_cis_element_analysis`; target_stages=[promoter]; parameters.promoter_length=2000; required_inputs=[genome_fasta,gff3,member_ids].",
  "required_dependencies": {},
  "required_params": {
    "promoter_length": 2000
  },
  "support_status": null,
  "target_stages": [
    "promoter"
  ]
}

Rationale

intent=`protein_property_analysis`; target_stages=[pep]; execution_plan=[pep]; required_inputs=[member_protein_fasta].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "pep"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "pep"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "easy"
  ],
  "task_id": "PGB-TU-006",
  "user_query": "“给我画这个基因家族的蛋白性质图。”上下文：成员蛋白 FASTA。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_protein_fasta"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "pep"
  ],
  "expected_execution_plan": [
    "pep"
  ],
  "expected_stages": [
    "pep"
  ],
  "forbidden_stages": [],
  "intent": "protein_property_analysis",
  "must_include": [],
  "must_not_do": [
    "不要误判为 expression 或 ppi",
    "不要声称能完成湿实验定位。"
  ],
  "parameters": {},
  "rationale": "intent=`protein_property_analysis`; target_stages=[pep]; execution_plan=[pep]; required_inputs=[member_protein_fasta].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "pep"
  ]
}

Rationale

intent=`multi_stage_gene_family_analysis`; species=maize; family=AP2-ERF; target_stages=[identify,domain,motif]; execution_plan=[identify,domain,motif].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "AP2-ERF",
    "params": {
      "geneFamily": "AP2-ERF",
      "species": "maize"
    },
    "species": "maize",
    "stages": [
      "identify",
      "domain",
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "AP2-ERF",
      "params": {
        "geneFamily": "AP2-ERF",
        "species": "maize"
      },
      "species": "maize",
      "stages": [
        "identify",
        "domain",
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-007",
  "user_query": "“先鉴定玉米 AP2-ERF，再做结构域和 motif。”上下文：有玉米蛋白组和 AP2 HMM。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "AP2-ERF",
    "species": "maize"
  },
  "evidence": [],
  "execution_plan": [
    "identify",
    "domain",
    "motif"
  ],
  "expected_execution_plan": [
    "identify",
    "domain",
    "motif"
  ],
  "expected_stages": [
    "identify",
    "domain",
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "multi_stage_gene_family_analysis",
  "must_include": [],
  "must_not_do": [
    "不要把 domain 和 motif 合并成一个阶段",
    "不要跳过 identify。"
  ],
  "parameters": {
    "geneFamily": "AP2-ERF",
    "species": "maize"
  },
  "rationale": "intent=`multi_stage_gene_family_analysis`; species=maize; family=AP2-ERF; target_stages=[identify,domain,motif]; execution_plan=[identify,domain,motif].",
  "required_dependencies": {},
  "required_params": {
    "geneFamily": "AP2-ERF",
    "species": "maize"
  },
  "support_status": null,
  "target_stages": [
    "identify",
    "domain",
    "motif"
  ]
}

Rationale

intent=`reference_based_subfamily_classification`; entities: target_species=Fagopyrum_tataricum, reference_species=Arabidopsis; target_stages=[reference_classification or subfamily_confirm]; execution_plan includes reference classification/phylogeny as needed.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "SPL",
    "params": {
      "species": "Fagopyrum_tataricum"
    },
    "species": "Arabidopsis",
    "stages": [
      "reference_classification or subfamily_confirm"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "SPL",
      "params": {
        "species": "Fagopyrum_tataricum"
      },
      "species": "Arabidopsis",
      "stages": [
        "reference_classification or subfamily_confirm"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-008",
  "user_query": "“我有 FtSPL 成员，想和拟南芥 SPL 一起分亚家族。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "SPL",
    "species": "Arabidopsis"
  },
  "evidence": [],
  "execution_plan": [
    "reference_classification or subfamily_confirm"
  ],
  "expected_execution_plan": [
    "reference_classification or subfamily_confirm"
  ],
  "expected_stages": [
    "reference_classification or subfamily_confirm"
  ],
  "forbidden_stages": [],
  "intent": "reference_based_subfamily_classification",
  "must_include": [],
  "must_not_do": [
    "不要把它当普通 identify",
    "不要发明亚家族结果。"
  ],
  "parameters": {
    "species": "Fagopyrum_tataricum"
  },
  "rationale": "intent=`reference_based_subfamily_classification`; entities: target_species=Fagopyrum_tataricum, reference_species=Arabidopsis; target_stages=[reference_classification or subfamily_confirm]; execution_plan includes reference classification/phylogeny as needed.",
  "required_dependencies": {},
  "required_params": {
    "species": "Fagopyrum_tataricum"
  },
  "support_status": null,
  "target_stages": [
    "reference_classification or subfamily_confirm"
  ]
}

Rationale

intent=`interspecies_synteny_analysis`; species=[rice,Arabidopsis]; target_stages=[interspecies_synteny]; execution_plan=[interspecies_synteny] after prerequisites.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": "rice",
    "stages": [
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": "rice",
      "stages": [
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-009",
  "user_query": "“比较水稻和拟南芥这个家族的共线性。”上下文：两个物种 GFF、CDS/pep 或等价 collinearity 输入。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "species": "rice"
  },
  "evidence": [],
  "execution_plan": [
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "interspecies_synteny"
  ],
  "expected_stages": [
    "interspecies_synteny"
  ],
  "forbidden_stages": [
    "synteny"
  ],
  "intent": "interspecies_synteny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要只运行 synteny（种内）",
    "不要把 karyotype 当主要目标。"
  ],
  "parameters": {},
  "rationale": "intent=`interspecies_synteny_analysis`; species=[rice,Arabidopsis]; target_stages=[interspecies_synteny]; execution_plan=[interspecies_synteny] after prerequisites.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "interspecies_synteny"
  ]
}

Rationale

intent=`gene_family_basic_workflow`; target_stages=[identify,phylogeny,domain,motif,structure,promoter,synteny,combo]; execution_plan respects dependencies and available inputs.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "identify",
      "phylogeny",
      "domain",
      "motif",
      "structure",
      "promoter",
      "synteny",
      "combo"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "identify",
        "phylogeny",
        "domain",
        "motif",
        "structure",
        "promoter",
        "synteny",
        "combo"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-010",
  "user_query": "“我想做 WRKY 的完整基础分析，后面要放论文图。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "expected_execution_plan": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "expected_stages": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "forbidden_stages": [],
  "intent": "gene_family_basic_workflow",
  "must_include": [],
  "must_not_do": [
    "不要自动加入 RNA-seq、PPI、Ka/Ks，除非用户要求或有数据。"
  ],
  "parameters": {},
  "rationale": "intent=`gene_family_basic_workflow`; target_stages=[identify,phylogeny,domain,motif,structure,promoter,synteny,combo]; execution_plan respects dependencies and available inputs.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ]
}

Rationale

intent=`gene_expression_visualization`; species=rice; id_type=gene_id; target_stages=[expression]; execution_plan=[expression]; required_inputs=[gene_ids,expression_matrix].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {
      "species": "rice"
    },
    "species": "rice",
    "stages": [
      "expression"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {
        "species": "rice"
      },
      "species": "rice",
      "stages": [
        "expression"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-011",
  "user_query": "“这些 Os01g0100100 这样的 ID 想做表达热图。”上下文：有表达矩阵。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "gene_ids",
      "expression_matrix"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "id_type": "gene_id",
    "species": "rice"
  },
  "evidence": [],
  "execution_plan": [
    "expression"
  ],
  "expected_execution_plan": [
    "expression"
  ],
  "expected_stages": [
    "expression"
  ],
  "forbidden_stages": [],
  "intent": "gene_expression_visualization",
  "must_include": [],
  "must_not_do": [
    "不要误判为 rnaseq_de",
    "不要要求 raw FASTQ。"
  ],
  "parameters": {
    "species": "rice"
  },
  "rationale": "intent=`gene_expression_visualization`; species=rice; id_type=gene_id; target_stages=[expression]; execution_plan=[expression]; required_inputs=[gene_ids,expression_matrix].",
  "required_dependencies": {},
  "required_params": {
    "species": "rice"
  },
  "support_status": null,
  "target_stages": [
    "expression"
  ]
}

Rationale

intent=`rnaseq_sample_quality_assessment`; target_stages=[rnaseq_sample_qc]; execution_plan=[rnaseq_sample_qc]; required_inputs=[count_matrix,sample_info].

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_sample_qc"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_sample_qc"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-012",
  "user_query": "“我上传了 sample_info 和 count matrix，想检查样本质量。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "count_matrix",
      "sample_info"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_sample_qc"
  ],
  "expected_execution_plan": [
    "rnaseq_sample_qc"
  ],
  "expected_stages": [
    "rnaseq_sample_qc"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_sample_quality_assessment",
  "must_include": [],
  "must_not_do": [
    "不要运行 raw read QC",
    "不要直接做 DE。"
  ],
  "parameters": {},
  "rationale": "intent=`rnaseq_sample_quality_assessment`; target_stages=[rnaseq_sample_qc]; execution_plan=[rnaseq_sample_qc]; required_inputs=[count_matrix,sample_info].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_sample_qc"
  ]
}

Rationale

intent=`rnaseq_pipeline`; target_stages=[rnaseq_qc,rnaseq_align,rnaseq_quant,rnaseq_de]; execution_plan follows QC -> align -> quant -> DE; enrich optional after DE.

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_qc",
      "rnaseq_align",
      "rnaseq_quant",
      "rnaseq_de"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_qc",
        "rnaseq_align",
        "rnaseq_quant",
        "rnaseq_de"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-013",
  "user_query": "“我有 raw fastq 和参考基因组，想从头跑 RNA-seq。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de"
  ],
  "expected_execution_plan": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de"
  ],
  "expected_stages": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_pipeline",
  "must_include": [],
  "must_not_do": [
    "不要跳过 QC/align 直接 DE",
    "不要把 expression stage 当 raw RNA-seq pipeline。"
  ],
  "parameters": {},
  "rationale": "intent=`rnaseq_pipeline`; target_stages=[rnaseq_qc,rnaseq_align,rnaseq_quant,rnaseq_de]; execution_plan follows QC -> align -> quant -> DE; enrich optional after DE.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de"
  ]
}

Rationale

intent=`candidate_family_recommendation`; stress=salt; analysis_input=DEG_result; target_stages=[rnaseq_candidate_family]; execution_plan=[rnaseq_candidate_family]; may use enrichment/family annotation evidence.

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {
      "analysis_input": "DEG_result"
    },
    "species": null,
    "stages": [
      "rnaseq_candidate_family"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {
        "analysis_input": "DEG_result"
      },
      "species": null,
      "stages": [
        "rnaseq_candidate_family"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-014",
  "user_query": "“我有盐胁迫 RNA-seq 差异基因结果，想找一个转录因子家族继续做下游分析。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [
    "rnaseq_candidate_family"
  ],
  "expected_execution_plan": [
    "rnaseq_candidate_family"
  ],
  "expected_stages": [
    "rnaseq_candidate_family"
  ],
  "forbidden_stages": [],
  "intent": "candidate_family_recommendation",
  "must_include": [],
  "must_not_do": [
    "不要直接选择 WRKY/MYB 等家族",
    "不要说已发现机制。"
  ],
  "parameters": {
    "analysis_input": "DEG_result"
  },
  "rationale": "intent=`candidate_family_recommendation`; stress=salt; analysis_input=DEG_result; target_stages=[rnaseq_candidate_family]; execution_plan=[rnaseq_candidate_family]; may use enrichment/family annotation evidence.",
  "required_dependencies": {},
  "required_params": {
    "analysis_input": "DEG_result"
  },
  "support_status": null,
  "target_stages": [
    "rnaseq_candidate_family"
  ]
}

Rationale

intent=`evidence_constrained_family_relevance_assessment`; target_stages=[expression,evidence_search] if data available; boundary_action=request_expression_or_literature_context.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "expression",
      "evidence_search"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "expression",
        "evidence_search"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "medium"
  ],
  "task_id": "PGB-TU-015",
  "user_query": "“帮我看一下这个家族是不是和干旱有关。”上下文：未给表达矩阵或文献证据。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "request_expression_or_literature_context.",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "expression",
    "evidence_search"
  ],
  "expected_execution_plan": [
    "expression",
    "evidence_search"
  ],
  "expected_stages": [
    "expression",
    "evidence_search"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_family_relevance_assessment",
  "must_include": [],
  "must_not_do": [
    "不要仅凭家族名断言干旱相关",
    "不要声称已验证。"
  ],
  "parameters": {},
  "rationale": "intent=`evidence_constrained_family_relevance_assessment`; target_stages=[expression,evidence_search] if data available; boundary_action=request_expression_or_literature_context.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "expression",
    "evidence_search"
  ]
}

Rationale

intent=`gene_family_expansion_assessment`; target_stages=[synteny,phylogeny,possibly_kaks]; boundary_action=request genome annotation/duplication or comparative data.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny",
      "phylogeny",
      "possibly_kaks"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny",
        "phylogeny",
        "possibly_kaks"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "hard"
  ],
  "task_id": "PGB-TU-016",
  "user_query": "“这个基因家族扩张了吗？”上下文：只有成员 ID 列表。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "request",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny",
    "phylogeny",
    "possibly_kaks"
  ],
  "expected_execution_plan": [
    "synteny",
    "phylogeny",
    "possibly_kaks"
  ],
  "expected_stages": [
    "synteny",
    "phylogeny",
    "possibly_kaks"
  ],
  "forbidden_stages": [],
  "intent": "gene_family_expansion_assessment",
  "must_include": [],
  "must_not_do": [
    "不要只根据成员数量断言扩张",
    "不要发明复制事件。"
  ],
  "parameters": {},
  "rationale": "intent=`gene_family_expansion_assessment`; target_stages=[synteny,phylogeny,possibly_kaks]; boundary_action=request genome annotation/duplication or comparative data.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "synteny",
    "phylogeny",
    "possibly_kaks"
  ]
}

Rationale

intent=`literature_methodology_to_workflow_alignment`; target_stages=[identify,motif,synteny]; qRT-PCR marked external_or_manual; evidence required.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "identify",
      "motif",
      "synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "identify",
        "motif",
        "synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "hard"
  ],
  "task_id": "PGB-TU-017",
  "user_query": "“把这篇论文的方法转成 bbav5 能跑的流程。”上下文：方法片段包含 HMMER、MEME、MCScanX、qRT-PCR。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "mark_external_or_manual",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "identify",
    "motif",
    "synteny"
  ],
  "expected_execution_plan": [
    "identify",
    "motif",
    "synteny"
  ],
  "expected_stages": [
    "identify",
    "motif",
    "synteny"
  ],
  "forbidden_stages": [],
  "intent": "literature_methodology_to_workflow_alignment",
  "must_include": [],
  "must_not_do": [
    "不要把 qRT-PCR 写成 bbav5 可自动执行阶段",
    "不要丢失证据句。"
  ],
  "parameters": {},
  "rationale": "intent=`literature_methodology_to_workflow_alignment`; target_stages=[identify,motif,synteny]; qRT-PCR marked external_or_manual; evidence required.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "external_or_manual",
  "target_stages": [
    "identify",
    "motif",
    "synteny"
  ]
}

Rationale

intent=`boundary_check_for_identification`; target_stages=[identify] requested but can_execute=false; boundary_action=request protein_fasta/genome and HMM/seed sequences; DEG can support candidate selection, not identification.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "identify"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "identify"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "hard"
  ],
  "task_id": "PGB-TU-018",
  "user_query": "“我只有水稻盐胁迫 DEG 表，能不能直接鉴定 WRKY 家族？”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "request",
  "can_execute": false,
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [
    "identify"
  ],
  "expected_execution_plan": [
    "identify"
  ],
  "expected_stages": [
    "identify"
  ],
  "forbidden_stages": [],
  "intent": "boundary_check_for_identification",
  "must_include": [],
  "must_not_do": [
    "不要用 DEG 表直接鉴定家族成员。"
  ],
  "parameters": {},
  "rationale": "intent=`boundary_check_for_identification`; target_stages=[identify] requested but can_execute=false; boundary_action=request protein_fasta/genome and HMM/seed sequences; DEG can support candidate selection, not identification.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify"
  ]
}

Rationale

intent=`literature_case_replay_scope_check`; target_stages=computational subset only; boundary_action=separate bbav5-supported workflow from wet/manual steps; can_execute depends on inputs.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "SPL",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "SPL",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "hard"
  ],
  "task_id": "PGB-TU-019",
  "user_query": "“帮我复现 FtSPL 论文所有实验。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "separate",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "SPL"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "literature_case_replay_scope_check",
  "must_include": [],
  "must_not_do": [
    "不要承诺完整复现论文",
    "不要把湿实验或人工数据库下载说成已自动完成。"
  ],
  "parameters": {},
  "rationale": "intent=`literature_case_replay_scope_check`; target_stages=computational subset only; boundary_action=separate bbav5-supported workflow from wet/manual steps; can_execute depends on inputs.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

intent=`candidate_gene_evidence_review`; target_stages=[biological_interpretation,evidence_constraint]; boundary_action=limited_candidate_support_request_literature_or_validation.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "biological_interpretation",
      "evidence_constraint"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "task_understanding",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "biological_interpretation",
        "evidence_constraint"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "task_understanding",
  "scoring_profile": [
    "intent",
    "entities",
    "target_stage",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "task_understanding",
    "hard"
  ],
  "task_id": "PGB-TU-020",
  "user_query": "“这批结果够不够写候选基因结论？”上下文：有 DE 表、domain 表、promoter 元件、无文献。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "limited_candidate_support_request_literature_or_validation.",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "biological_interpretation",
    "evidence_constraint"
  ],
  "expected_execution_plan": [
    "biological_interpretation",
    "evidence_constraint"
  ],
  "expected_stages": [
    "biological_interpretation",
    "evidence_constraint"
  ],
  "forbidden_stages": [],
  "intent": "candidate_gene_evidence_review",
  "must_include": [],
  "must_not_do": [
    "不要写成机制已证明",
    "不要忽略缺失文献/实验验证。"
  ],
  "parameters": {},
  "rationale": "intent=`candidate_gene_evidence_review`; target_stages=[biological_interpretation,evidence_constraint]; boundary_action=limited_candidate_support_request_literature_or_validation.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "biological_interpretation",
    "evidence_constraint"
  ]
}

Rationale

target_stages=[identify,phylogeny,domain,motif,structure,promoter,synteny,combo]; execution_plan starts with identify; use member outputs for downstream stages.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "identify",
      "phylogeny",
      "domain",
      "motif",
      "structure",
      "promoter",
      "synteny",
      "combo"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "identify",
        "phylogeny",
        "domain",
        "motif",
        "structure",
        "promoter",
        "synteny",
        "combo"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-001",
  "user_query": "“完成水稻 WRKY 基础分析。”上下文：蛋白组、GFF3、基因组 FASTA、WRKY HMM。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "expected_execution_plan": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "expected_stages": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ],
  "forbidden_stages": [],
  "intent": "combo_analysis",
  "must_include": [],
  "must_not_do": [
    "不要先跑 motif/domain",
    "不要加入 RNA-seq。"
  ],
  "parameters": {},
  "rationale": "target_stages=[identify,phylogeny,domain,motif,structure,promoter,synteny,combo]; execution_plan starts with identify; use member outputs for downstream stages.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "promoter",
    "synteny",
    "combo"
  ]
}

Rationale

target_stages=[domain]; execution_plan=[domain]; selected_inputs=[member_protein_fasta]; expected_outputs include domain table/plot/manifest.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-002",
  "user_query": "“已有 `final_members.pep.fa`，只做 domain 检查。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_protein_fasta"
    ],
    "selected_inputs": [
      "member_protein_fasta"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "domain_analysis",
  "must_include": [],
  "must_not_do": [
    "不要重新 identify",
    "不要用 nucleotide FASTA。"
  ],
  "parameters": {},
  "rationale": "target_stages=[domain]; execution_plan=[domain]; selected_inputs=[member_protein_fasta]; expected_outputs include domain table/plot/manifest.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain"
  ]
}

Rationale

target_stages=[structure]; execution_plan=[structure]; selected_inputs=[member_ids,gff3].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "structure"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "structure"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-003",
  "user_query": "“已有 `final_members.id` 和 GFF3，画 gene structure。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_ids",
      "gff3"
    ],
    "selected_inputs": [
      "member_ids",
      "gff3"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "structure"
  ],
  "expected_execution_plan": [
    "structure"
  ],
  "expected_stages": [
    "structure"
  ],
  "forbidden_stages": [],
  "intent": "structure_analysis",
  "must_include": [],
  "must_not_do": [
    "不要要求 HMM",
    "不要把结构域图当 gene structure。"
  ],
  "parameters": {},
  "rationale": "target_stages=[structure]; execution_plan=[structure]; selected_inputs=[member_ids,gff3].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "structure"
  ]
}

Rationale

target_stages=[pep]; execution_plan=[pep]; selected_inputs=[member_protein_fasta]; output=protein property/subcellular or TM-related summaries as supported by pep stage.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "pep"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "pep"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-004",
  "user_query": "“给成员蛋白预测理化性质和跨膜信息。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_protein_fasta"
    ],
    "selected_inputs": [
      "member_protein_fasta"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "pep"
  ],
  "expected_execution_plan": [
    "pep"
  ],
  "expected_stages": [
    "pep"
  ],
  "forbidden_stages": [
    "ppi"
  ],
  "intent": "pep_analysis",
  "must_include": [],
  "must_not_do": [
    "不要声称实验定位",
    "不要调用 PPI。"
  ],
  "parameters": {},
  "rationale": "target_stages=[pep]; execution_plan=[pep]; selected_inputs=[member_protein_fasta]; output=protein property/subcellular or TM-related summaries as supported by pep stage.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "pep"
  ]
}

Rationale

target_stages=[combo]; execution_plan=[combo]; required_inputs=[tree,motif_results,structure_results] or available final artifacts.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "combo"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "combo"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-005",
  "user_query": "“我想把 tree、motif、structure 合成一张组合图。”上下文：已有对应阶段产物。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "tree",
      "motif_results",
      "structure_results"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "combo"
  ],
  "expected_execution_plan": [
    "combo"
  ],
  "expected_stages": [
    "combo"
  ],
  "forbidden_stages": [],
  "intent": "combo_analysis",
  "must_include": [],
  "must_not_do": [
    "不要重新运行上游分析",
    "不要缺上游图时假装可合成。"
  ],
  "parameters": {},
  "rationale": "target_stages=[combo]; execution_plan=[combo]; required_inputs=[tree,motif_results,structure_results] or available final artifacts.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "combo"
  ]
}

Rationale

target_stages=[synteny]; execution_plan=[synteny]; distinguish intra-species synteny.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "easy"
  ],
  "task_id": "PGB-EP-006",
  "user_query": "“做种内共线性。”上下文：单物种 genome/GFF/protein 准备齐。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny"
  ],
  "expected_execution_plan": [
    "synteny"
  ],
  "expected_stages": [
    "synteny"
  ],
  "forbidden_stages": [
    "interspecies_synteny"
  ],
  "intent": "synteny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要运行 interspecies_synteny",
    "不要把 karyotype 当共线性主体。"
  ],
  "parameters": {},
  "rationale": "target_stages=[synteny]; execution_plan=[synteny]; distinguish intra-species synteny.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "synteny"
  ]
}

Rationale

target_stages=[motif]; execution_plan=[identify,motif]; identify output `member_protein_fasta` feeds motif.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-007",
  "user_query": "“还没鉴定 OsWRKY 成员，但想直接看 motif 分布。”上下文：水稻蛋白组和 WRKY HMM。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "identify",
    "motif"
  ],
  "expected_execution_plan": [
    "identify",
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "motif_analysis",
  "must_include": [],
  "must_not_do": [
    "不要在全蛋白组上跑 MEME",
    "不要拒绝而不提出依赖补全。"
  ],
  "parameters": {},
  "rationale": "target_stages=[motif]; execution_plan=[identify,motif]; identify output `member_protein_fasta` feeds motif.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "motif"
  ]
}

Rationale

target_stages=[tree_plot]; execution_plan=[phylogeny,tree_plot] or phylogeny with tree plot output; selected_input=member_protein_fasta.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "tree_plot"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "tree_plot"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-008",
  "user_query": "“没做 phylogeny，但想画 circular tree。”上下文：成员蛋白 FASTA。"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny",
    "tree_plot"
  ],
  "expected_execution_plan": [
    "phylogeny",
    "tree_plot"
  ],
  "expected_stages": [
    "tree_plot"
  ],
  "forbidden_stages": [],
  "intent": "tree_plot_analysis",
  "must_include": [],
  "must_not_do": [
    "不要画空树",
    "不要把 tree_plot 当无需树文件的独立分析。"
  ],
  "parameters": {},
  "rationale": "target_stages=[tree_plot]; execution_plan=[phylogeny,tree_plot] or phylogeny with tree plot output; selected_input=member_protein_fasta.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "tree_plot"
  ]
}

Rationale

target_stages=[promoter]; execution_plan=[]; boundary_action=request_genome_fasta; required_inputs=[member_ids,gff3,genome_fasta].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "promoter"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "promoter"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-009",
  "user_query": "“我要 promoter 图，但只有成员 ID 和 GFF，没有 genome FASTA。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "member_ids",
      "gff3",
      "genome_fasta"
    ],
    "selected_inputs": []
  },
  "boundary_action": "request_genome_fasta",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "promoter"
  ],
  "expected_execution_plan": [
    "promoter"
  ],
  "expected_stages": [
    "promoter"
  ],
  "forbidden_stages": [],
  "intent": "promoter_analysis",
  "must_include": [],
  "must_not_do": [
    "不要发明 promoter sequence",
    "不要用 GFF 直接提取序列。"
  ],
  "parameters": {},
  "rationale": "target_stages=[promoter]; execution_plan=[]; boundary_action=request_genome_fasta; required_inputs=[member_ids,gff3,genome_fasta].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "promoter"
  ]
}

Rationale

target_stages=[interspecies_synteny]; execution_plan=[interspecies_synteny]; required_inputs include both species annotations/sequences; outputs include pairwise synteny artifacts.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-010",
  "user_query": "“有两个物种的 GFF 和蛋白序列，想做种间共线性。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "interspecies_synteny"
  ],
  "expected_stages": [
    "interspecies_synteny"
  ],
  "forbidden_stages": [],
  "intent": "interspecies_synteny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要只做单物种 synteny",
    "不要默认有 layout_mark 产物。"
  ],
  "parameters": {},
  "rationale": "target_stages=[interspecies_synteny]; execution_plan=[interspecies_synteny]; required_inputs include both species annotations/sequences; outputs include pairwise synteny artifacts.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "interspecies_synteny"
  ]
}

Rationale

target_stages=[promoter]; parameters.promoter_length=1500; required_inputs=[genome_fasta,gff3,member_ids].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {
      "promoter_length": 1500
    },
    "species": null,
    "stages": [
      "promoter"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {
        "promoter_length": 1500
      },
      "species": null,
      "stages": [
        "promoter"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-011",
  "user_query": "“上游 1500 bp 做顺式元件分析。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "genome_fasta",
      "gff3",
      "member_ids"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "promoter"
  ],
  "expected_execution_plan": [
    "promoter"
  ],
  "expected_stages": [
    "promoter"
  ],
  "forbidden_stages": [],
  "intent": "promoter_analysis",
  "must_include": [],
  "must_not_do": [
    "不要覆盖成 2000 bp 默认值",
    "不要忽略用户参数。"
  ],
  "parameters": {
    "promoter_length": 1500
  },
  "rationale": "target_stages=[promoter]; parameters.promoter_length=1500; required_inputs=[genome_fasta,gff3,member_ids].",
  "required_dependencies": {},
  "required_params": {
    "promoter_length": 1500
  },
  "support_status": null,
  "target_stages": [
    "promoter"
  ]
}

Rationale

target_stages=[motif]; parameters.max_motifs=10, motif_width=[6,200]; selected_input=member_protein_fasta.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {
      "max_motifs": 10
    },
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {
        "max_motifs": 10
      },
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-012",
  "user_query": "“MEME 最多找 10 个 motif，宽度 6-200。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "motif"
  ],
  "expected_execution_plan": [
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "motif_analysis",
  "must_include": [],
  "must_not_do": [
    "不要只记录 max_motifs",
    "不要改用不可追溯默认参数。"
  ],
  "parameters": {
    "max_motifs": 10
  },
  "rationale": "target_stages=[motif]; parameters.max_motifs=10, motif_width=[6,200]; selected_input=member_protein_fasta.",
  "required_dependencies": {},
  "required_params": {
    "max_motifs": 10
  },
  "support_status": null,
  "target_stages": [
    "motif"
  ]
}

Rationale

target_stages=[phylogeny]; execution_plan=[phylogeny]; parameters.tree_method=ML, bootstrap=1000 if supported; note bbav5 default may use MAFFT/IQ-TREE.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {
      "bootstrap": 1000,
      "tree_method": "ML"
    },
    "species": null,
    "stages": [
      "phylogeny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {
        "bootstrap": 1000,
        "tree_method": "ML"
      },
      "species": null,
      "stages": [
        "phylogeny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-013",
  "user_query": "“phylogeny 用 ML 方法和 1000 bootstrap。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny"
  ],
  "expected_execution_plan": [
    "phylogeny"
  ],
  "expected_stages": [
    "phylogeny"
  ],
  "forbidden_stages": [],
  "intent": "phylogeny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要写成 NJ/MEGA 默认",
    "不要忽略 bootstrap。"
  ],
  "parameters": {
    "bootstrap": 1000,
    "tree_method": "ML"
  },
  "rationale": "target_stages=[phylogeny]; execution_plan=[phylogeny]; parameters.tree_method=ML, bootstrap=1000 if supported; note bbav5 default may use MAFFT/IQ-TREE.",
  "required_dependencies": {},
  "required_params": {
    "bootstrap": 1000,
    "tree_method": "ML"
  },
  "support_status": null,
  "target_stages": [
    "phylogeny"
  ]
}

Rationale

target_stages=[identify]; parameters.hmmer.evalue=1e-5, blast.evalue=1e-5; required_inputs=[protein_fasta,hmm_or_seed].

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "MYB",
    "params": {},
    "species": null,
    "stages": [
      "identify"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "MYB",
      "params": {},
      "species": null,
      "stages": [
        "identify"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-014",
  "user_query": "“BLAST e-value 1e-5，HMMER e-value 1e-5 鉴定 MYB。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "protein_fasta",
      "hmm_or_seed"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "MYB"
  },
  "evidence": [],
  "execution_plan": [
    "identify"
  ],
  "expected_execution_plan": [
    "identify"
  ],
  "expected_stages": [
    "identify"
  ],
  "forbidden_stages": [],
  "intent": "identify_analysis",
  "must_include": [],
  "must_not_do": [
    "不要改阈值",
    "不要只跑 BLAST 忽略 HMMER。"
  ],
  "parameters": {},
  "rationale": "target_stages=[identify]; parameters.hmmer.evalue=1e-5, blast.evalue=1e-5; required_inputs=[protein_fasta,hmm_or_seed].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify"
  ]
}

Rationale

target_stages=[expression]; execution_plan=[expression]; selected_inputs=[expression_matrix,gene_ids]; no raw RNA-seq stages.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "expression"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "expression"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-015",
  "user_query": "“表达矩阵已经有了，只画候选基因热图。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "expression_matrix",
      "gene_ids"
    ],
    "selected_inputs": [
      "expression_matrix",
      "gene_ids"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "expression"
  ],
  "expected_execution_plan": [
    "expression"
  ],
  "expected_stages": [
    "expression"
  ],
  "forbidden_stages": [
    "rnaseq_qc"
  ],
  "intent": "expression_analysis",
  "must_include": [],
  "must_not_do": [
    "不要运行 rnaseq_qc/align/quant",
    "不要要求 FASTQ。"
  ],
  "parameters": {},
  "rationale": "target_stages=[expression]; execution_plan=[expression]; selected_inputs=[expression_matrix,gene_ids]; no raw RNA-seq stages.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "expression"
  ]
}

Rationale

target_stages=[rnaseq_qc,rnaseq_align,rnaseq_quant,rnaseq_de,rnaseq_enrich]; execution_plan follows dependency; enrich after DE; use rnaseq env in implementation.

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_qc",
      "rnaseq_align",
      "rnaseq_quant",
      "rnaseq_de",
      "rnaseq_enrich"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_qc",
        "rnaseq_align",
        "rnaseq_quant",
        "rnaseq_de",
        "rnaseq_enrich"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-016",
  "user_query": "“有 raw RNA-seq FASTQ，要做差异表达和 GO/KEGG。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de",
    "rnaseq_enrich"
  ],
  "expected_execution_plan": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de",
    "rnaseq_enrich"
  ],
  "expected_stages": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de",
    "rnaseq_enrich"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_enrich_analysis",
  "must_include": [],
  "must_not_do": [
    "不要直接 enrich",
    "不要用 expression stage 代替 DE。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_qc,rnaseq_align,rnaseq_quant,rnaseq_de,rnaseq_enrich]; execution_plan follows dependency; enrich after DE; use rnaseq env in implementation.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_qc",
    "rnaseq_align",
    "rnaseq_quant",
    "rnaseq_de",
    "rnaseq_enrich"
  ]
}

Rationale

target_stages=[rnaseq_de]; execution_plan=[rnaseq_de]; required_inputs=[count_matrix,sample_info,contrast]; can skip align/quant because count matrix exists.

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_de"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_de"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-017",
  "user_query": "“有 count matrix 和 sample_info，跳过比对，直接 DESeq2 差异分析。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "count_matrix",
      "sample_info",
      "contrast"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_de"
  ],
  "expected_execution_plan": [
    "rnaseq_de"
  ],
  "expected_stages": [
    "rnaseq_de"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_de_analysis",
  "must_include": [],
  "must_not_do": [
    "不要强制 raw pipeline",
    "不要丢失 contrast。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_de]; execution_plan=[rnaseq_de]; required_inputs=[count_matrix,sample_info,contrast]; can skip align/quant because count matrix exists.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_de"
  ]
}

Rationale

target_stages=[rnaseq_candidate_family]; execution_plan=[rnaseq_candidate_family]; selected_inputs=[DEG_table,enrichment_results,family_annotation]; output=candidate family ranking.

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_candidate_family"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_candidate_family"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-018",
  "user_query": "“基于差异基因、GO/KEGG 和 Pfam family annotation 推荐候选家族。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "DEG_table",
      "enrichment_results",
      "family_annotation"
    ],
    "selected_inputs": [
      "DEG_table",
      "enrichment_results",
      "family_annotation"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_candidate_family"
  ],
  "expected_execution_plan": [
    "rnaseq_candidate_family"
  ],
  "expected_stages": [
    "rnaseq_candidate_family"
  ],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要直接启动 identify",
    "不要随机选择家族。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_candidate_family]; execution_plan=[rnaseq_candidate_family]; selected_inputs=[DEG_table,enrichment_results,family_annotation]; output=candidate family ranking.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_candidate_family"
  ]
}

Rationale

target_stages=[rnaseq_wgcna]; execution_plan=[rnaseq_wgcna]; required_inputs=[expression_matrix,trait_table or sample_info].

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_wgcna"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_wgcna"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-019",
  "user_query": "“做 WGCNA 找盐胁迫相关模块。”上下文：表达矩阵和表型/时间点信息。"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "expression_matrix",
      "trait_table or sample_info"
    ],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [
    "rnaseq_wgcna"
  ],
  "expected_execution_plan": [
    "rnaseq_wgcna"
  ],
  "expected_stages": [
    "rnaseq_wgcna"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_wgcna_analysis",
  "must_include": [],
  "must_not_do": [
    "不要把 WGCNA 当 GO enrichment",
    "不要没有样本设计就运行。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_wgcna]; execution_plan=[rnaseq_wgcna]; required_inputs=[expression_matrix,trait_table or sample_info].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_wgcna"
  ]
}

Rationale

target_stages=[rnaseq_fgsea]; execution_plan=[rnaseq_fgsea]; selected_inputs=[ranked_gene_list,gene_sets].

Input context

{
  "intent": {
    "analysis_type": "rnaseq",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_fgsea"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "rnaseq",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_fgsea"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "medium"
  ],
  "task_id": "PGB-EP-020",
  "user_query": "“做 FGSEA，已有 ranked gene list 和 gene sets。”"
}

Full gold JSON

{
  "analysis_type": "rnaseq",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [
      "ranked_gene_list",
      "gene_sets"
    ],
    "selected_inputs": [
      "ranked_gene_list",
      "gene_sets"
    ]
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_fgsea"
  ],
  "expected_execution_plan": [
    "rnaseq_fgsea"
  ],
  "expected_stages": [
    "rnaseq_fgsea"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_fgsea_analysis",
  "must_include": [],
  "must_not_do": [
    "不要要求 raw count",
    "不要把 FGSEA 与 over-representation GO 混淆。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_fgsea]; execution_plan=[rnaseq_fgsea]; selected_inputs=[ranked_gene_list,gene_sets].",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_fgsea"
  ]
}

Rationale

target_stages=[phylogeny]; execution_plan=[phylogeny]; support_status=supported_alternative; replacement=MAFFT/IQ-TREE or supported phylogeny tools; evidence required.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "phylogeny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "phylogeny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-021",
  "user_query": "“一篇文献用 ClustalW + MEGA NJ 1000 bootstrap 建树，转成 bbav5 流程。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny"
  ],
  "expected_execution_plan": [
    "phylogeny"
  ],
  "expected_stages": [
    "phylogeny"
  ],
  "forbidden_stages": [],
  "intent": "phylogeny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要说 bbav5 原样运行 MEGA",
    "不要丢掉替代说明。"
  ],
  "parameters": {},
  "rationale": "target_stages=[phylogeny]; execution_plan=[phylogeny]; support_status=supported_alternative; replacement=MAFFT/IQ-TREE or supported phylogeny tools; evidence required.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "phylogeny"
  ]
}

Rationale

target_stages=[pep]; execution_plan=[pep]; support_status=supported_alternative; note pep covers computational protein-property/localization-style outputs, not experimental localization.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "pep"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "pep"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-022",
  "user_query": "“文献用 ExPASy ProtParam 和 Cell-PLoc 做蛋白性质/定位，映射到系统。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "pep"
  ],
  "expected_execution_plan": [
    "pep"
  ],
  "expected_stages": [
    "pep"
  ],
  "forbidden_stages": [],
  "intent": "pep_analysis",
  "must_include": [],
  "must_not_do": [
    "不要标为 unsupported",
    "不要声称等同 Cell-PLoc 原工具。"
  ],
  "parameters": {},
  "rationale": "target_stages=[pep]; execution_plan=[pep]; support_status=supported_alternative; note pep covers computational protein-property/localization-style outputs, not experimental localization.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "pep"
  ]
}

Rationale

target_stages=[] for executable bbav5; support_status=external_or_manual; execution_plan should keep qRT-PCR as manual validation note after expression/RNA-seq evidence if relevant.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "expression"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "expression"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-023",
  "user_query": "“文献用 qRT-PCR 验证表达，请加入 workflow。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "mark_external_or_manual",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "expression"
  ],
  "expected_execution_plan": [
    "expression"
  ],
  "expected_stages": [
    "expression"
  ],
  "forbidden_stages": [],
  "intent": "expression_analysis",
  "must_include": [],
  "must_not_do": [
    "不要创建 qRT-PCR CLI 阶段",
    "不要说系统自动完成实验验证。"
  ],
  "parameters": {},
  "rationale": "target_stages=[] for executable bbav5; support_status=external_or_manual; execution_plan should keep qRT-PCR as manual validation note after expression/RNA-seq evidence if relevant.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "external_or_manual",
  "target_stages": [
    "expression"
  ]
}

Rationale

target_stages=[ortholog,kaks,interspecies_synteny]; execution_plan validates ortholog/pair inputs before kaks; interspecies synteny uses pairwise species inputs.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "ortholog",
      "kaks",
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "ortholog",
        "kaks",
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-024",
  "user_query": "“我想做 ortholog、Ka/Ks、种间共线性一起解释复制与选择压力。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "ortholog",
    "kaks",
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "ortholog",
    "kaks",
    "interspecies_synteny"
  ],
  "expected_stages": [
    "ortholog",
    "kaks",
    "interspecies_synteny"
  ],
  "forbidden_stages": [],
  "intent": "interspecies_synteny_analysis",
  "must_include": [],
  "must_not_do": [
    "不要先跑 Ka/Ks 无同源基因对",
    "不要把 Ka/Ks 当核心 13 阶段之一。"
  ],
  "parameters": {},
  "rationale": "target_stages=[ortholog,kaks,interspecies_synteny]; execution_plan validates ortholog/pair inputs before kaks; interspecies synteny uses pairwise species inputs.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "ortholog",
    "kaks",
    "interspecies_synteny"
  ]
}

Rationale

target_stages=[identify per species,phylogeny,domain,motif,structure,interspecies_synteny,karyotype/combo as needed]; execution_plan must be per-species for identify and pairwise for interspecies.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": "SPL",
    "params": {},
    "species": null,
    "stages": [
      "identify per species",
      "phylogeny",
      "domain",
      "motif",
      "structure",
      "interspecies_synteny",
      "karyotype/combo as needed"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": "SPL",
      "params": {},
      "species": null,
      "stages": [
        "identify per species",
        "phylogeny",
        "domain",
        "motif",
        "structure",
        "interspecies_synteny",
        "karyotype/combo as needed"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-025",
  "user_query": "“跨 5 个物种做 SPL 比较基因组分析。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "SPL"
  },
  "evidence": [],
  "execution_plan": [
    "identify per species",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "interspecies_synteny",
    "karyotype/combo as needed"
  ],
  "expected_execution_plan": [
    "identify per species",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "interspecies_synteny",
    "karyotype/combo as needed"
  ],
  "expected_stages": [
    "identify per species",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "interspecies_synteny",
    "karyotype/combo as needed"
  ],
  "forbidden_stages": [],
  "intent": "karyotype/combo as needed_analysis",
  "must_include": [],
  "must_not_do": [
    "不要把 5 物种混成一个 species",
    "不要忽略 per-species outputs。"
  ],
  "parameters": {},
  "rationale": "target_stages=[identify per species,phylogeny,domain,motif,structure,interspecies_synteny,karyotype/combo as needed]; execution_plan must be per-species for identify and pairwise for interspecies.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify per species",
    "phylogeny",
    "domain",
    "motif",
    "structure",
    "interspecies_synteny",
    "karyotype/combo as needed"
  ]
}

Rationale

target_stages=[motif,domain,structure]; execution_plan uses confirmed subfamily scope/member_set; selected_inputs must be scoped member FASTA/IDs.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "motif",
      "domain",
      "structure"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "motif",
        "domain",
        "structure"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-026",
  "user_query": "“亚家族确认后，只对某个 subfamily 画 motif/domain/structure。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "motif",
    "domain",
    "structure"
  ],
  "expected_execution_plan": [
    "motif",
    "domain",
    "structure"
  ],
  "expected_stages": [
    "motif",
    "domain",
    "structure"
  ],
  "forbidden_stages": [],
  "intent": "structure_analysis",
  "must_include": [],
  "must_not_do": [
    "不要回到 all members",
    "不要丢失 scope/member_set。"
  ],
  "parameters": {},
  "rationale": "target_stages=[motif,domain,structure]; execution_plan uses confirmed subfamily scope/member_set; selected_inputs must be scoped member FASTA/IDs.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "motif",
    "domain",
    "structure"
  ]
}

Rationale

target_stages=[motif]; execution_plan=[]; boundary_action=artifact_missing_blocker; require actual member_protein_fasta.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-027",
  "user_query": "“已有 manifest 显示 identify 成功，但 artifacts 缺 member protein FASTA，能继续 motif 吗？”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "artifact_missing_blocker",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "motif"
  ],
  "expected_execution_plan": [
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "motif_analysis",
  "must_include": [],
  "must_not_do": [
    "不要只看 status=success 就继续",
    "不要发明 artifact。"
  ],
  "parameters": {},
  "rationale": "target_stages=[motif]; execution_plan=[]; boundary_action=artifact_missing_blocker; require actual member_protein_fasta.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "motif"
  ]
}

Rationale

target_stages=[rnaseq_enrich or rnaseq_enrich_v2 depending input]; execution_plan requires DEG/gene list plus gene2go/gene2ko mapping or supported annotation tables.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "rnaseq_enrich or rnaseq_enrich_v2 depending input"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "rnaseq_enrich or rnaseq_enrich_v2 depending input"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-028",
  "user_query": "“已有 annotation 输出，想做 GO/KEGG 富集。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "rnaseq_enrich or rnaseq_enrich_v2 depending input"
  ],
  "expected_execution_plan": [
    "rnaseq_enrich or rnaseq_enrich_v2 depending input"
  ],
  "expected_stages": [
    "rnaseq_enrich or rnaseq_enrich_v2 depending input"
  ],
  "forbidden_stages": [],
  "intent": "rnaseq_enrich or rnaseq_enrich_v2 depending input_analysis",
  "must_include": [],
  "must_not_do": [
    "不要没有背景集/映射就运行",
    "不要假设所有物种都有 OrgDb。"
  ],
  "parameters": {},
  "rationale": "target_stages=[rnaseq_enrich or rnaseq_enrich_v2 depending input]; execution_plan requires DEG/gene list plus gene2go/gene2ko mapping or supported annotation tables.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "rnaseq_enrich or rnaseq_enrich_v2 depending input"
  ]
}

Rationale

target_stages=[ppi]; execution_plan=[ppi] only if supported species/network source or protein IDs available; boundary_action=request species and accepted identifier mapping.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "ppi"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "ppi"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-029",
  "user_query": "“只给了 gene IDs，要做 PPI 网络。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "request",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "ppi"
  ],
  "expected_execution_plan": [
    "ppi"
  ],
  "expected_stages": [
    "ppi"
  ],
  "forbidden_stages": [],
  "intent": "ppi_analysis",
  "must_include": [],
  "must_not_do": [
    "不要跨物种硬套 Arabidopsis STRING",
    "不要发明互作。"
  ],
  "parameters": {},
  "rationale": "target_stages=[ppi]; execution_plan=[ppi] only if supported species/network source or protein IDs available; boundary_action=request species and accepted identifier mapping.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "ppi"
  ]
}

Rationale

target_stages=[summary]; execution_plan=[summary] if upstream manifests/artifacts exist; required_inputs=upstream manifests/artifacts; output=integrated summary/report.

Input context

{
  "intent": {
    "analysis_type": "gene_family",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "summary"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "execution_planning",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "gene_family",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "summary"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "execution_planning",
  "scoring_profile": [
    "target_stage",
    "execution_plan",
    "parameters",
    "artifacts",
    "boundary"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "execution_planning",
    "hard"
  ],
  "task_id": "PGB-EP-030",
  "user_query": "“把 identify、domain、motif、structure、promoter、synteny 的结果整理成 summary。”"
}

Full gold JSON

{
  "analysis_type": "gene_family",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "summary"
  ],
  "expected_execution_plan": [
    "summary"
  ],
  "expected_stages": [
    "summary"
  ],
  "forbidden_stages": [],
  "intent": "summary_analysis",
  "must_include": [],
  "must_not_do": [
    "不要重跑所有阶段",
    "不要汇总不存在的结果。"
  ],
  "parameters": {},
  "rationale": "target_stages=[summary]; execution_plan=[summary] if upstream manifests/artifacts exist; required_inputs=upstream manifests/artifacts; output=integrated summary/report.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "summary"
  ]
}

Rationale

map to identify; evidence claim source=literature_excerpt; support_status=supported_default or supported_alternative depending exact tool/catalog.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "identify"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“HMMER was used to identify family members.”问：对应系统哪个阶段？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "identify"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "easy"
  ],
  "task_id": "PGB-EC-001",
  "user_query": "文献片段：“HMMER was used to identify family members.”问：对应系统哪个阶段？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "identify"
  ],
  "expected_execution_plan": [
    "identify"
  ],
  "expected_stages": [
    "identify"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要映射到 domain-only",
    "不要省略证据句。"
  ],
  "parameters": {},
  "rationale": "map to identify; evidence claim source=literature_excerpt; support_status=supported_default or supported_alternative depending exact tool/catalog.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_default",
  "target_stages": [
    "identify"
  ]
}

Rationale

map to motif; tool=MEME; support_status=supported_default if motif stage uses MEME-compatible workflow.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“MEME was used to identify conserved motifs.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "easy"
  ],
  "task_id": "PGB-EC-002",
  "user_query": "文献片段：“MEME was used to identify conserved motifs.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "motif"
  ],
  "expected_execution_plan": [
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要映射到 domain",
    "不要把 motif 解释成 cis-element。"
  ],
  "parameters": {},
  "rationale": "map to motif; tool=MEME; support_status=supported_default if motif stage uses MEME-compatible workflow.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_default",
  "target_stages": [
    "motif"
  ]
}

Rationale

map to synteny or interspecies_synteny depending context; if species context absent, ask whether intra- or interspecies; evidence-supported.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny",
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“MCScanX was used for collinearity analysis.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny",
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "easy"
  ],
  "task_id": "PGB-EC-003",
  "user_query": "文献片段：“MCScanX was used for collinearity analysis.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny",
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "synteny",
    "interspecies_synteny"
  ],
  "expected_stages": [
    "synteny",
    "interspecies_synteny"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要默认 interspecies",
    "不要忽略上下文不足。"
  ],
  "parameters": {},
  "rationale": "map to synteny or interspecies_synteny depending context; if species context absent, ask whether intra- or interspecies; evidence-supported.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "synteny",
    "interspecies_synteny"
  ]
}

Rationale

map to phylogeny; tools=[ClustalW,MEGA]; parameters=[NJ,1000 bootstrap]; support_status=supported_alternative.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "phylogeny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“Clustal W alignment and MEGA NJ tree with 1000 bootstrap.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "phylogeny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "easy"
  ],
  "task_id": "PGB-EC-004",
  "user_query": "文献片段：“Clustal W alignment and MEGA NJ tree with 1000 bootstrap.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny"
  ],
  "expected_execution_plan": [
    "phylogeny"
  ],
  "expected_stages": [
    "phylogeny"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要说 bbav5 默认就是 MEGA",
    "不要丢参数。"
  ],
  "parameters": {},
  "rationale": "map to phylogeny; tools=[ClustalW,MEGA]; parameters=[NJ,1000 bootstrap]; support_status=supported_alternative.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "phylogeny"
  ]
}

Rationale

classify as external_or_manual; may relate to expression evidence but not executable bbav5 stage.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "expression"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“qRT-PCR was performed to validate expression.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "expression"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "easy"
  ],
  "task_id": "PGB-EC-005",
  "user_query": "文献片段：“qRT-PCR was performed to validate expression.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "mark_external_or_manual",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "expression"
  ],
  "expected_execution_plan": [
    "expression"
  ],
  "expected_stages": [
    "expression"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要映射为 rnaseq_de 或 expression plotting",
    "不要声称自动验证。"
  ],
  "parameters": {},
  "rationale": "classify as external_or_manual; may relate to expression evidence but not executable bbav5 stage.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "external_or_manual",
  "target_stages": [
    "expression"
  ]
}

Rationale

extract three method observations and map to identify, motif, synteny/interspecies_synteny; include one evidence item per mapping.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "identify",
      "motif",
      "synteny",
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段同时包含 HMMER、MEME、MCScanX。",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "identify",
        "motif",
        "synteny",
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-006",
  "user_query": "文献片段同时包含 HMMER、MEME、MCScanX。"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "identify",
    "motif",
    "synteny",
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "identify",
    "motif",
    "synteny",
    "interspecies_synteny"
  ],
  "expected_stages": [
    "identify",
    "motif",
    "synteny",
    "interspecies_synteny"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要合成一个 vague “analysis” 阶段",
    "不要漏工具。"
  ],
  "parameters": {},
  "rationale": "extract three method observations and map to identify, motif, synteny/interspecies_synteny; include one evidence item per mapping.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "identify",
    "motif",
    "synteny",
    "interspecies_synteny"
  ]
}

Rationale

map to pep; support_status=supported_alternative; evidence supports protein property/localization prediction only.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "pep"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“ExPASy calculated MW, pI, instability index; Cell-PLoc predicted localization.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "pep"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-007",
  "user_query": "文献片段：“ExPASy calculated MW, pI, instability index; Cell-PLoc predicted localization.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "pep"
  ],
  "expected_execution_plan": [
    "pep"
  ],
  "expected_stages": [
    "pep"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要写成实验定位",
    "不要标为 promoter。"
  ],
  "parameters": {},
  "rationale": "map to pep; support_status=supported_alternative; evidence supports protein property/localization prediction only.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "pep"
  ]
}

Rationale

map to domain; parameter.cd_search_evalue=0.01; support_status=supported_alternative if bbav5 uses Pfam/HMMER domain workflow.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“NCBI CD-Search predicted conserved domains with e-value 0.01.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-008",
  "user_query": "文献片段：“NCBI CD-Search predicted conserved domains with e-value 0.01.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要丢阈值",
    "不要把 CD-Search 原样当默认工具。"
  ],
  "parameters": {},
  "rationale": "map to domain; parameter.cd_search_evalue=0.01; support_status=supported_alternative if bbav5 uses Pfam/HMMER domain workflow.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "domain"
  ]
}

Rationale

map duplication to synteny/interspecies_synteny as relevant; Ka/Ks to kaks; evidence should separate duplication and selection pressure claims.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny",
      "interspecies_synteny",
      "kaks"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“Tandem and segmental duplication events were identified and Ka/Ks ratios calculated.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny",
        "interspecies_synteny",
        "kaks"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-009",
  "user_query": "文献片段：“Tandem and segmental duplication events were identified and Ka/Ks ratios calculated.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny",
    "interspecies_synteny",
    "kaks"
  ],
  "expected_execution_plan": [
    "synteny",
    "interspecies_synteny",
    "kaks"
  ],
  "expected_stages": [
    "synteny",
    "interspecies_synteny",
    "kaks"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把 Ka/Ks 作为证明功能的直接证据",
    "不要混淆两类输出。"
  ],
  "parameters": {},
  "rationale": "map duplication to synteny/interspecies_synteny as relevant; Ka/Ks to kaks; evidence should separate duplication and selection pressure claims.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "synteny",
    "interspecies_synteny",
    "kaks"
  ]
}

Rationale

map to promoter; parameter.promoter_length=2000; support_status=supported_default/alternative depending implementation.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {
      "promoter_length": 2000
    },
    "species": null,
    "stages": [
      "promoter"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段：“PlantCARE was used to analyze 2 kb upstream promoters.”",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {
        "promoter_length": 2000
      },
      "species": null,
      "stages": [
        "promoter"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-010",
  "user_query": "文献片段：“PlantCARE was used to analyze 2 kb upstream promoters.”"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "promoter"
  ],
  "expected_execution_plan": [
    "promoter"
  ],
  "expected_stages": [
    "promoter"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要默认 1500 bp",
    "不要说 promoter result implies functional validation。"
  ],
  "parameters": {
    "promoter_length": 2000
  },
  "rationale": "map to promoter; parameter.promoter_length=2000; support_status=supported_default/alternative depending implementation.",
  "required_dependencies": {},
  "required_params": {
    "promoter_length": 2000
  },
  "support_status": "supported_default",
  "target_stages": [
    "promoter"
  ]
}

Rationale

conclusion=候选支持但机制未证明; evidence combines family-level literature + dataset expression; boundary_action=candidate_supported_not_mechanism_proven.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "expression"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "RAG 证据 A：WRKY family often participates in abiotic stress; 证据 B：OsWRKY12 is upregulated in this dataset; 问能否说 OsWRKY12 抗盐？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "expression"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-011",
  "user_query": "RAG 证据 A：WRKY family often participates in abiotic stress; 证据 B：OsWRKY12 is upregulated in this dataset; 问能否说 OsWRKY12 抗盐？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "candidate_supported_not_mechanism_proven.",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [
    "expression"
  ],
  "expected_execution_plan": [
    "expression"
  ],
  "expected_stages": [
    "expression"
  ],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要写“已证明抗盐基因”",
    "不要只引用 family-level evidence。"
  ],
  "parameters": {},
  "rationale": "conclusion=候选支持但机制未证明; evidence combines family-level literature + dataset expression; boundary_action=candidate_supported_not_mechanism_proven.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "expression"
  ]
}

Rationale

answer=no; evidence=artifact missing despite manifest declaration; boundary_action=artifact_validation_required.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "结果片段：manifest status=success, artifacts includes `motif_plot`, file exists=false。问能否引用 motif 图？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-012",
  "user_query": "结果片段：manifest status=success, artifacts includes `motif_plot`, file exists=false。问能否引用 motif 图？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "artifact_validation_required.",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要仅凭 status=success 引用图",
    "不要生成不存在路径。"
  ],
  "parameters": {},
  "rationale": "answer=no; evidence=artifact missing despite manifest declaration; boundary_action=artifact_validation_required.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=不建议作为高置信成员；需要 HMM/domain validation; evidence=domain table; boundary_action=flag_for_manual_review or remove/low confidence.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "结果片段：domain table 显示候选蛋白无核心 WRKY domain，仅有低复杂度 motif。问是否保留为 WRKY 成员？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-013",
  "user_query": "结果片段：domain table 显示候选蛋白无核心 WRKY domain，仅有低复杂度 motif。问是否保留为 WRKY 成员？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "flag_for_manual_review",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要只凭 motif 保留",
    "不要说一定是假阳性。"
  ],
  "parameters": {},
  "rationale": "answer=不建议作为高置信成员；需要 HMM/domain validation; evidence=domain table; boundary_action=flag_for_manual_review or remove/low confidence.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain"
  ]
}

Rationale

answer=表达支持 stress responsiveness，但需 family membership/domain/annotation evidence; boundary_action=claim_limited.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "domain",
      "annotation"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "结果片段：DE 表显示 log2FC=2.0, FDR=0.001；无 domain/annotation。问是否能作为家族候选基因？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "domain",
        "annotation"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "medium"
  ],
  "task_id": "PGB-EC-014",
  "user_query": "结果片段：DE 表显示 log2FC=2.0, FDR=0.001；无 domain/annotation。问是否能作为家族候选基因？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "claim_limited.",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "domain",
    "annotation"
  ],
  "expected_execution_plan": [
    "domain",
    "annotation"
  ],
  "expected_stages": [
    "domain",
    "annotation"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把表达证据当家族成员证据。"
  ],
  "parameters": {},
  "rationale": "answer=表达支持 stress responsiveness，但需 family membership/domain/annotation evidence; boundary_action=claim_limited.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain",
    "annotation"
  ]
}

Rationale

MCScanX/synteny computation maps to synteny/interspecies_synteny if supported; TBtools visualization may be supported_alternative/partially_supported or external visualization; record support_status separately.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny",
      "interspecies_synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献方法使用 TBtools 可视化 MCScanX 结果。问如何对齐 bbav5？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny",
        "interspecies_synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-015",
  "user_query": "文献方法使用 TBtools 可视化 MCScanX 结果。问如何对齐 bbav5？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny",
    "interspecies_synteny"
  ],
  "expected_execution_plan": [
    "synteny",
    "interspecies_synteny"
  ],
  "expected_stages": [
    "synteny",
    "interspecies_synteny"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把 TBtools GUI 全部写成 bbav5 默认",
    "不要丢可视化边界。"
  ],
  "parameters": {},
  "rationale": "MCScanX/synteny computation maps to synteny/interspecies_synteny if supported; TBtools visualization may be supported_alternative/partially_supported or external visualization; record support_status separately.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "synteny",
    "interspecies_synteny"
  ]
}

Rationale

classify data retrieval as external/input preparation unless corresponding fetch stage is configured; phylogeny is executable after sequences exist; evidence separates download and tree.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "phylogeny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献用 TAIR、RGAP 下载参考蛋白序列用于建树。问是否 bbav5 可自动完成？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "phylogeny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-016",
  "user_query": "文献用 TAIR、RGAP 下载参考蛋白序列用于建树。问是否 bbav5 可自动完成？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny"
  ],
  "expected_execution_plan": [
    "phylogeny"
  ],
  "expected_stages": [
    "phylogeny"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把数据库下载当 phylogeny",
    "不要声称已自动获取。"
  ],
  "parameters": {},
  "rationale": "classify data retrieval as external/input preparation unless corresponding fetch stage is configured; phylogeny is executable after sequences exist; evidence separates download and tree.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "phylogeny"
  ]
}

Rationale

boundary_action=refuse_unsupported_wording; acceptable wording=候选/推测/需要验证; evidence insufficiency noted.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "用户要求“没有文献也写 OsWRKY12 是已验证盐胁迫调控因子”。",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-017",
  "user_query": "用户要求“没有文献也写 OsWRKY12 是已验证盐胁迫调控因子”。"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "refuse_unsupported_wording",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要按用户要求夸大",
    "不要编造引用。"
  ],
  "parameters": {},
  "rationale": "boundary_action=refuse_unsupported_wording; acceptable wording=候选/推测/需要验证; evidence insufficiency noted.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "unsupported",
  "target_stages": []
}

Rationale

select WRKY stress literature + project DE result; reject unrelated SPL evidence; evidence_faithfulness requires relevance.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "RAG 返回 3 条证据：一条 WRKY 盐胁迫综述、一条无关 SPL 花发育、一条本项目 DE 结果。问选择哪些支持 OsWRKY 盐胁迫候选。",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-018",
  "user_query": "RAG 返回 3 条证据：一条 WRKY 盐胁迫综述、一条无关 SPL 花发育、一条本项目 DE 结果。问选择哪些支持 OsWRKY 盐胁迫候选。"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要引用无关 SPL 证据",
    "不要把综述当基因特异验证。"
  ],
  "parameters": {},
  "rationale": "select WRKY stress literature + project DE result; reject unrelated SPL evidence; evidence_faithfulness requires relevance.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

classify as biological finding evidence, not method observation unless method sentence present; may support interpretation with caveat.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "文献片段来自 Results 而非 Methods，描述“gene family expanded via WGD”。问能否作为 workflow 方法证据？",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-019",
  "user_query": "文献片段来自 Results 而非 Methods，描述“gene family expanded via WGD”。问能否作为 workflow 方法证据？"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要当成可执行方法步骤",
    "不要抽取不存在工具。"
  ],
  "parameters": {},
  "rationale": "classify as biological finding evidence, not method observation unless method sentence present; may support interpretation with caveat.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

extract identify observation for BLAST/member search; extract domain validation for CD-Search/domain; support_status=supported_alternative/default as appropriate; preserve e-value thresholds.

Input context

{
  "intent": {
    "analysis_type": "evidence_constraint",
    "gene_family": "BBX",
    "params": {},
    "species": null,
    "stages": [
      "identify",
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "evidence_constraint",
  "context": {
    "files": [],
    "literature_excerpt": "对齐 gold：BBX 文献用 BLAST + NCBI CD-Search；bbav5 catalog identify 默认 HMMER+BLAST，domain 另有 domain stage。问如何拆分。",
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": null
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "evidence_constraint",
      "gene_family": "BBX",
      "params": {},
      "species": null,
      "stages": [
        "identify",
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "evidence_constraint",
  "scoring_profile": [
    "target_stage",
    "evidence",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "evidence_constraint",
    "hard"
  ],
  "task_id": "PGB-EC-020",
  "user_query": "对齐 gold：BBX 文献用 BLAST + NCBI CD-Search；bbav5 catalog identify 默认 HMMER+BLAST，domain 另有 domain stage。问如何拆分。"
}

Full gold JSON

{
  "analysis_type": "evidence_constraint",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "BBX"
  },
  "evidence": [],
  "execution_plan": [
    "identify",
    "domain"
  ],
  "expected_execution_plan": [
    "identify",
    "domain"
  ],
  "expected_stages": [
    "identify",
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "evidence_constrained_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把两个步骤混为 identify-only",
    "不要丢 CD-Search e-value。"
  ],
  "parameters": {},
  "rationale": "extract identify observation for BLAST/member search; extract domain validation for CD-Search/domain; support_status=supported_alternative/default as appropriate; preserve e-value thresholds.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": "supported_alternative",
  "target_stages": [
    "identify",
    "domain"
  ]
}

Rationale

answer=支持高置信 WRKY 成员； evidence=domain/HMM hit; still require ID consistency and redundancy check.

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：候选蛋白有完整 WRKY domain，HMMER e-value 1e-20。问是否支持 WRKY 成员。"
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "easy"
  ],
  "task_id": "PGB-BI-001",
  "user_query": "结果：候选蛋白有完整 WRKY domain，HMMER e-value 1e-20。问是否支持 WRKY 成员。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要说功能已验证",
    "不要忽略基因模型质量。"
  ],
  "parameters": {},
  "rationale": "answer=支持高置信 WRKY 成员； evidence=domain/HMM hit; still require ID consistency and redundancy check.",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain"
  ]
}

Rationale

answer=提示家族内保守结构特征，支持成员一致性/亚群特征；需结合 domain/phylogeny。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "phylogeny",
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：MEME motif 图显示多数成员共享 motif1/2/3。问能说明什么？"
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "phylogeny",
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "easy"
  ],
  "task_id": "PGB-BI-002",
  "user_query": "结果：MEME motif 图显示多数成员共享 motif1/2/3。问能说明什么？"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny",
    "domain"
  ],
  "expected_execution_plan": [
    "phylogeny",
    "domain"
  ],
  "expected_stages": [
    "phylogeny",
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要说 motif 直接证明功能。"
  ],
  "parameters": {},
  "rationale": "answer=提示家族内保守结构特征，支持成员一致性/亚群特征；需结合 domain/phylogeny。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "phylogeny",
    "domain"
  ]
}

Rationale

answer=提示可能受 ABA/胁迫相关调控，属于候选调控线索；需表达/实验验证。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：启动子含 ABRE、DRE 元件。问能否说明胁迫响应。"
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "easy"
  ],
  "task_id": "PGB-BI-003",
  "user_query": "结果：启动子含 ABRE、DRE 元件。问能否说明胁迫响应。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要写成已证明响应。"
  ],
  "parameters": {},
  "rationale": "answer=提示可能受 ABA/胁迫相关调控，属于候选调控线索；需表达/实验验证。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=支持同源关系/亚家族归类假设；需要 bootstrap/support 和 domain/motif一致性。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：系统发育树中目标基因与拟南芥已注释同源基因聚在同一 clade。"
  },
  "difficulty": "easy",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "easy"
  ],
  "task_id": "PGB-BI-004",
  "user_query": "结果：系统发育树中目标基因与拟南芥已注释同源基因聚在同一 clade。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要仅凭单棵树推断完全相同功能。"
  ],
  "parameters": {},
  "rationale": "answer=支持同源关系/亚家族归类假设；需要 bootstrap/support 和 domain/motif一致性。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain"
  ]
}

Rationale

answer=低置信或需人工复核；不能作为高置信家族成员；建议检查蛋白长度、domain、annotation。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "domain",
      "annotation"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：某候选成员只有 MEME motif 命中，无 Pfam/HMM domain。问是否保留。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "domain",
        "annotation"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-005",
  "user_query": "结果：某候选成员只有 MEME motif 命中，无 Pfam/HMM domain。问是否保留。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "domain",
    "annotation"
  ],
  "expected_execution_plan": [
    "domain",
    "annotation"
  ],
  "expected_stages": [
    "domain",
    "annotation"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要只凭 motif 保留为正式成员。"
  ],
  "parameters": {},
  "rationale": "answer=低置信或需人工复核；不能作为高置信家族成员；建议检查蛋白长度、domain、annotation。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain",
    "annotation"
  ]
}

Rationale

answer=可能提示亚群特异结构差异或序列发散；需检查 motif detection 参数、序列完整性、tree support。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "motif"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：某 clade 缺失一个保守 motif，但 domain 完整。问如何解释。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "motif"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-006",
  "user_query": "结果：某 clade 缺失一个保守 motif，但 domain 完整。问如何解释。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "motif"
  ],
  "expected_execution_plan": [
    "motif"
  ],
  "expected_stages": [
    "motif"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要断言功能丧失。"
  ],
  "parameters": {},
  "rationale": "answer=可能提示亚群特异结构差异或序列发散；需检查 motif detection 参数、序列完整性、tree support。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "motif"
  ]
}

Rationale

answer=支持 tandem duplication 作为家族扩张证据之一；需定义距离/间隔规则并结合 synteny。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "synteny"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：一组基因在同一染色体相邻位置形成串联重复。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "synteny"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-007",
  "user_query": "结果：一组基因在同一染色体相邻位置形成串联重复。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "synteny"
  ],
  "expected_execution_plan": [
    "synteny"
  ],
  "expected_stages": [
    "synteny"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要说一定由串联重复导致全部扩张。"
  ],
  "parameters": {},
  "rationale": "answer=支持 tandem duplication 作为家族扩张证据之一；需定义距离/间隔规则并结合 synteny。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "synteny"
  ]
}

Rationale

answer=支持 segmental/WGD-related duplication hypothesis; 可与 Ka/Ks/物种背景结合。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：跨染色体共线 blocks 中出现多个家族成员。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-008",
  "user_query": "结果：跨染色体共线 blocks 中出现多个家族成员。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要把共线性等同功能冗余。"
  ],
  "parameters": {},
  "rationale": "answer=支持 segmental/WGD-related duplication hypothesis; 可与 Ka/Ks/物种背景结合。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=通常提示纯化选择；需说明模型/同源对质量，不能单独证明功能保守。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：Ka/Ks < 1。问说明什么。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-009",
  "user_query": "结果：Ka/Ks < 1。问说明什么。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要写成正选择",
    "不要忽略 dS 饱和风险。"
  ],
  "parameters": {},
  "rationale": "answer=通常提示纯化选择；需说明模型/同源对质量，不能单独证明功能保守。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=提供互作假设/优先级线索；若来自预测数据库需标明预测性质并需实验验证。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：PPI 网络显示候选蛋白与多个胁迫相关蛋白互作预测。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-010",
  "user_query": "结果：PPI 网络显示候选蛋白与多个胁迫相关蛋白互作预测。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要说实验证实互作。"
  ],
  "parameters": {},
  "rationale": "answer=提供互作假设/优先级线索；若来自预测数据库需标明预测性质并需实验验证。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=可作为优先候选；证据=差异表达+成员质量；不能证明调控机制。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：OsWRKY12 盐胁迫 6h/24h 上调，FDR<0.01，有完整 domain。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-011",
  "user_query": "结果：OsWRKY12 盐胁迫 6h/24h 上调，FDR<0.01，有完整 domain。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "claim_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要声称抗盐功能已验证。"
  ],
  "parameters": {},
  "rationale": "answer=可作为优先候选；证据=差异表达+成员质量；不能证明调控机制。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=支持该家族值得下游分析/候选排序；需考虑背景家族大小和多重检验。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：某家族 DEG 占比高，GO 富集在 response to salt stress。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-012",
  "user_query": "结果：某家族 DEG 占比高，GO 富集在 response to salt stress。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要说整个家族都抗盐。"
  ],
  "parameters": {},
  "rationale": "answer=支持该家族值得下游分析/候选排序；需考虑背景家族大小和多重检验。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=支持 NAC 作为候选家族或模块关联线索；需结合 DEG、domain、literature evidence。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "NAC",
    "params": {},
    "species": null,
    "stages": [
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：WGCNA 模块与盐处理时间显著相关，模块中富集 NAC 成员。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "NAC",
      "params": {},
      "species": null,
      "stages": [
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-013",
  "user_query": "结果：WGCNA 模块与盐处理时间显著相关，模块中富集 NAC 成员。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "NAC",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [
    "domain"
  ],
  "expected_execution_plan": [
    "domain"
  ],
  "expected_stages": [
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把相关性解释为因果。"
  ],
  "parameters": {},
  "rationale": "answer=支持 NAC 作为候选家族或模块关联线索；需结合 DEG、domain、literature evidence。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "domain"
  ]
}

Rationale

answer=支持 WRKY 相关基因集整体上调/富集；需说明 gene set 来源和排序统计。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：FGSEA 显示 WRKY target gene set NES>0, FDR<0.05。"
  },
  "difficulty": "medium",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "medium"
  ],
  "task_id": "PGB-BI-014",
  "user_query": "结果：FGSEA 显示 WRKY target gene set NES>0, FDR<0.05。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要替代单基因验证",
    "不要忽略 FDR。"
  ],
  "parameters": {},
  "rationale": "answer=支持 WRKY 相关基因集整体上调/富集；需说明 gene set 来源和排序统计。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=通常优先 B，因多证据一致且成员质量更可靠；A 可列为低置信或复核对象。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：候选 A 上调强但 domain 不完整；候选 B 上调中等、domain 完整、有共线性和 ABRE。问优先哪个。"
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "hard"
  ],
  "task_id": "PGB-BI-015",
  "user_query": "结果：候选 A 上调强但 domain 不完整；候选 B 上调中等、domain 完整、有共线性和 ABRE。问优先哪个。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要只按 log2FC 排名。"
  ],
  "parameters": {},
  "rationale": "answer=通常优先 B，因多证据一致且成员质量更可靠；A 可列为低置信或复核对象。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=亚家族分类需标记低置信；建议补充 motif/domain/参考集或重新建树。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "SPL",
    "params": {},
    "species": null,
    "stages": [
      "motif",
      "domain"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：reference_classification 将目标基因分到 SPL9/15-like，但树 bootstrap 低。"
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "SPL",
      "params": {},
      "species": null,
      "stages": [
        "motif",
        "domain"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "hard"
  ],
  "task_id": "PGB-BI-016",
  "user_query": "结果：reference_classification 将目标基因分到 SPL9/15-like，但树 bootstrap 低。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "SPL"
  },
  "evidence": [],
  "execution_plan": [
    "motif",
    "domain"
  ],
  "expected_execution_plan": [
    "motif",
    "domain"
  ],
  "expected_stages": [
    "motif",
    "domain"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要给确定分类",
    "不要忽略 bootstrap。"
  ],
  "parameters": {},
  "rationale": "answer=亚家族分类需标记低置信；建议补充 motif/domain/参考集或重新建树。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "motif",
    "domain"
  ]
}

Rationale

answer=可提出广义 abiotic stress 候选，但盐胁迫特异证据主要来自本数据；文献支撑需谨慎。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": "WRKY",
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：RNA-seq 显示 WRKY 家族响应盐胁迫，但文献主要支持干旱响应。"
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": "WRKY",
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "hard"
  ],
  "task_id": "PGB-BI-017",
  "user_query": "结果：RNA-seq 显示 WRKY 家族响应盐胁迫，但文献主要支持干旱响应。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {
    "gene_family": "WRKY",
    "stress": "salt"
  },
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "candidate_evidence_assessment",
  "must_include": [],
  "must_not_do": [
    "不要把干旱文献直接当盐胁迫证据。"
  ],
  "parameters": {},
  "rationale": "answer=可提出广义 abiotic stress 候选，但盐胁迫特异证据主要来自本数据；文献支撑需谨慎。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=不能仅凭数量；需 genome quality、annotation consistency、phylogeny/synteny/duplication evidence。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": [
      "phylogeny",
      "synteny",
      "annotation"
    ]
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：species A 有 80 个家族成员，species B 有 20 个。问能否说 A 家族扩张。"
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": [
        "phylogeny",
        "synteny",
        "annotation"
      ]
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "hard"
  ],
  "task_id": "PGB-BI-018",
  "user_query": "结果：species A 有 80 个家族成员，species B 有 20 个。问能否说 A 家族扩张。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "evidence_supported_or_limited",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [
    "phylogeny",
    "synteny",
    "annotation"
  ],
  "expected_execution_plan": [
    "phylogeny",
    "synteny",
    "annotation"
  ],
  "expected_stages": [
    "phylogeny",
    "synteny",
    "annotation"
  ],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要根据成员数直接断言扩张。"
  ],
  "parameters": {},
  "rationale": "answer=不能仅凭数量；需 genome quality、annotation consistency、phylogeny/synteny/duplication evidence。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": [
    "phylogeny",
    "synteny",
    "annotation"
  ]
}

Rationale

answer=报告 pairwise_success_rate 和失败边界；只解释成功物种对，失败原因放 limitation/supplement。

Input context

{
  "intent": {
    "analysis_type": "biological_interpretation",
    "gene_family": null,
    "params": {},
    "species": null,
    "stages": []
  },
  "job_id": "",
  "prefix": ""
}

Full task JSON

{
  "category": "biological_interpretation",
  "context": {
    "files": [],
    "literature_excerpt": null,
    "manifest_excerpt": null,
    "metadata": {},
    "result_excerpt": "结果：interspecies_synteny 只成功 6/10 物种对。问论文如何报告。"
  },
  "difficulty": "hard",
  "input_context": {
    "intent": {
      "analysis_type": "biological_interpretation",
      "gene_family": null,
      "params": {},
      "species": null,
      "stages": []
    },
    "job_id": "",
    "prefix": ""
  },
  "layer": "biological_interpretation",
  "scoring_profile": [
    "evidence",
    "biology",
    "boundary",
    "hallucination"
  ],
  "source_type": "synthetic_expert_curated",
  "tags": [
    "biological_interpretation",
    "hard"
  ],
  "task_id": "PGB-BI-019",
  "user_query": "结果：interspecies_synteny 只成功 6/10 物种对。问论文如何报告。"
}

Full gold JSON

{
  "analysis_type": "biological_interpretation",
  "artifacts": {
    "expected_outputs": [],
    "required_inputs": [],
    "selected_inputs": []
  },
  "boundary_action": "none",
  "can_execute": "depends_on_inputs",
  "entities": {},
  "evidence": [],
  "execution_plan": [],
  "expected_execution_plan": [],
  "expected_stages": [],
  "forbidden_stages": [],
  "intent": "biological_interpretation",
  "must_include": [],
  "must_not_do": [
    "不要隐瞒失败",
    "不要把不完整结果写成全物种结论。"
  ],
  "parameters": {},
  "rationale": "answer=报告 pairwise_success_rate 和失败边界；只解释成功物种对，失败原因放 limitation/supplement。",
  "required_dependencies": {},
  "required_params": {},
  "support_status": null,
  "target_stages": []
}

Rationale

answer=写“evidence-prioritized candidates”，说明权重和证据来源；需要敏感性/边界说明；不能称最终功能基因。