feat: add gene family result reports

2026-08-09 04:32:28 +08:00
parent 65f79207f0
commit cb122fd088
20 changed files with 3676 additions and 20 deletions
@@ -0,0 +1,226 @@
+#!/usr/bin/env python3
+"""Small real-data journey for the gene-family report module."""
+
+from __future__ import annotations
+
+import argparse
+import json
+from pathlib import Path
+import re
+import shutil
+import sys
+
+
+PROJECT_ROOT = Path(__file__).resolve().parents[2]
+if str(PROJECT_ROOT) not in sys.path:
+    sys.path.insert(0, str(PROJECT_ROOT))
+
+from apps.cli.run_gene_family_report import main as report_cli_main
+from gfa_core.mas.utils.intent_parser import parse_intent_from_query
+
+
+PREFIX = "Real.SPL"
+SCOPE_ID = "scope-real-spl"
+ANALYSIS_ROUTE = "post_subfamily_confirm"
+MEMBER_SET_ID = "members-real-spl"
+
+
+def _sequence_count(path: Path) -> int:
+    count = sum(line.startswith(">") for line in path.read_text(encoding="utf-8").splitlines())
+    if not count:
+        raise AssertionError("representative FASTA is empty")
+    return count
+
+
+def _write_identify_manifest(workspace: Path, fasta: Path, count: int) -> None:
+    final_dir = workspace / "final"
+    final_dir.mkdir(parents=True, exist_ok=True)
+    manifest = {
+        "manifest_version": "1.0",
+        "stage": "identify",
+        "status": "success",
+        "prefix": PREFIX,
+        "scope_id": SCOPE_ID,
+        "analysis_route": ANALYSIS_ROUTE,
+        "member_set_id": MEMBER_SET_ID,
+        "parameters": {
+            "scope_id": SCOPE_ID,
+            "analysis_route": ANALYSIS_ROUTE,
+            "member_set_id": MEMBER_SET_ID,
+        },
+        "metrics": {"member_count": count, "final_members": count, "total_proteins": count},
+        "artifacts": {
+            "final_members": {"path": str(fasta), "full_path": str(fasta), "type": "protein_fasta"}
+        },
+    }
+    (final_dir / f"{PREFIX}.identify.manifest.json").write_text(
+        json.dumps(manifest, ensure_ascii=False, indent=2) + "\n",
+        encoding="utf-8",
+    )
+
+
+def _run(workspace: Path, action: str):
+    return report_cli_main(
+        [
+            "--workspace", str(workspace),
+            "--prefix", PREFIX,
+            "--job-id", "representative-gene-family-report",
+            "--expected-stages", "identify",
+            "--stage-statuses-json", '{"identify":"success"}',
+            "--scope-id", SCOPE_ID,
+            "--analysis-route", ANALYSIS_ROUTE,
+            "--member-set-id", MEMBER_SET_ID,
+            "--action", action,
+            "--no-model",
+        ]
+    )
+
+
+def _stage_section(report: str, stage: str) -> str:
+    heading = re.search(rf"(?m)^## (?!#).*?\(`{re.escape(stage)}`\)", report)
+    if heading is None:
+        raise AssertionError(f"missing report chapter for {stage}")
+    next_heading = re.search(r"(?m)^## (?!#)", report[heading.end():])
+    end = heading.end() + next_heading.start() if next_heading else len(report)
+    return report[heading.start():end]
+
+
+def _subsection(report: str, number: str) -> str:
+    heading = re.search(rf"(?m)^### {re.escape(number)}(?:\s|$)", report)
+    if heading is None:
+        raise AssertionError(f"missing report subsection {number}")
+    next_heading = re.search(r"(?m)^#{2,3} ", report[heading.end():])
+    end = heading.end() + next_heading.start() if next_heading else len(report)
+    return report[heading.start():end]
+
+
+def _verify_real_job(workspace: Path, prefix: str) -> None:
+    final_dir = workspace / "final"
+    report_path = final_dir / f"{prefix}.gene_family_report.md"
+    manifest_path = final_dir / f"{prefix}.gene_family_report.manifest.json"
+    report = report_path.read_text(encoding="utf-8")
+    manifest = json.loads(manifest_path.read_text(encoding="utf-8"))
+    expected_stages = [
+        "identify",
+        "phylogeny",
+        "reference_classification",
+        "motif",
+        "domain",
+        "structure",
+        "combo",
+        "promoter",
+        "tfbs",
+        "synteny",
+        "karyotype",
+        "interspecies_synteny",
+        "pep",
+    ]
+
+    assert report.startswith(f"# {prefix} 基因家族分析结果解读报告\n")
+    for placeholder in ("【填写】", "未解析", "N/A", "None"):
+        assert placeholder not in report
+    assert "/home/xzq/" not in report
+    assert not re.search(r"(?m)^- `[a-zA-Z0-9_.-]+`:\s*", report)
+    assert manifest["parameters"]["expected_stages"] == expected_stages
+    assert manifest["metrics"]["included_stages"] == expected_stages
+    assert manifest["metrics"]["generation_mode"] == "fallback_script"
+    assert sorted(path.name for path in final_dir.glob(f"{prefix}.gene_family_report*")) == [
+        f"{prefix}.gene_family_report.manifest.json",
+        f"{prefix}.gene_family_report.md",
+    ]
+    for stage in expected_stages:
+        section = _stage_section(report, stage)
+        for chunk in re.split(r"(?m)(?=^### )", section)[1:]:
+            assert "\n".join(chunk.splitlines()[1:]).strip(), chunk.splitlines()[0]
+        assert "分析对象：" in section
+        assert "限制：" in section
+    assert report.count("本节解释仅限上述可追溯结果，不据此断言真实功能或因果关系。") <= 1
+    summary = _subsection(report, "1.3")
+    assert 6 <= summary.count("。") <= 10
+    assert re.search(r"目标物种.*24.*家族成员", summary)
+    assert re.search(r"形成\s*24\s*个\s*SPL家族", _subsection(report, "4.3"))
+    assert re.search(
+        r"使用\s*40\s*条序列.*比对长度为\s*1449", _stage_section(report, "phylogeny")
+    )
+    motif = _stage_section(report, "motif")
+    assert re.search(r"识别\s*10\s*个\s*motif", motif)
+    assert re.search(r"总位点数为\s*105", motif)
+    assert re.search(r"宽度范围为\s*27–138", motif)
+    assert re.search(r"将\s*24\s*个家族成员定位到\s*8\s*条染色体", _stage_section(report, "karyotype"))
+    assert re.search(
+        r"全基因组\s*block\s*521\s*个.*目标-参考家族.*?2\s*对",
+        _stage_section(report, "interspecies_synteny"),
+    )
+    assert re.search(r"覆盖\s*40\s*个蛋白，子任务成功数为\s*3/3", _stage_section(report, "pep"))
+    assert re.search(r"24\s*个目标成员.*8\s*个小组.*高信心\s*14.*中信心\s*7.*低信心\s*3", _subsection(report, "6.2"))
+    assert re.search(r"40/40.*SBP.*40.*Ig_SPL7.*12", _subsection(report, "8.1"))
+    assert re.search(r"24/40.*FtPinG0003446200\.01\.T01.*12454", _subsection(report, "9.1"))
+    assert "Motif、Domain、Structure" in _subsection(report, "10.1")
+    promoter = _subsection(report, "11.1")
+    assert re.search(r"23\s*个目标基因.*3699.*TATA-box.*1460", promoter)
+    assert "CAAT-box—MYB" in promoter and "FtPinG0006337300.01.T01" in promoter
+    tfbs = _subsection(report, "12.1")
+    assert re.search(r"24\s*条目标启动子.*3734.*AHYPO_015621.*153", tfbs)
+    assert re.search(r"背景启动子为\s*0.*不能.*富集", tfbs)
+    synteny = _subsection(report, "13.1")
+    assert re.search(r"全基因组.*316.*3267.*8\s*条染色体", synteny)
+    assert "源结果未提供家族专属复制基因对" in synteny
+    assert re.search(r"Nucleus\s*37.*Cytoplasm\s*2.*Chloroplast\s*1", _subsection(report, "16.4"))
+
+
+def main() -> int:
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--mode", choices=("data", "e2e"))
+    parser.add_argument("--workspace", type=Path)
+    parser.add_argument("--prefix")
+    args = parser.parse_args()
+
+    if args.workspace is not None:
+        if not args.prefix:
+            parser.error("--prefix is required with --workspace")
+        _verify_real_job(args.workspace, args.prefix)
+        print(json.dumps({"status": "passed", "workspace": str(args.workspace)}, ensure_ascii=False))
+        return 0
+    if args.mode is None:
+        parser.error("--mode is required when --workspace is not provided")
+
+    output_dir = PROJECT_ROOT / "test-results/gene-family-report" / args.mode
+    if output_dir.exists():
+        shutil.rmtree(output_dir)
+    workspace = output_dir / "job"
+    fasta = PROJECT_ROOT / "tests/ft_data/SPL_Ath.fasta"
+    count = _sequence_count(fasta)
+    _write_identify_manifest(workspace, fasta, count)
+
+    generated = _run(workspace, "generate")
+    report = generated.report_path.read_text(encoding="utf-8")
+    manifest = json.loads(generated.manifest_path.read_text(encoding="utf-8"))
+    assert generated.report_path.name == f"{PREFIX}.gene_family_report.md"
+    assert generated.manifest_path.name == f"{PREFIX}.gene_family_report.manifest.json"
+    assert "生成模式：候补脚本" in report
+    assert f"最终成员数：{count}" in report
+    assert "## 基因家族鉴定 (`identify`)" in report
+    assert "(`motif`)" not in report
+    assert manifest["scope_id"] == SCOPE_ID
+    assert manifest["analysis_route"] == ANALYSIS_ROUTE
+    assert manifest["member_set_id"] == MEMBER_SET_ID
+    assert manifest["metrics"]["included_stages"] == ["identify"]
+    assert set(manifest["artifacts"]) == {"gene_family_report", "gene_family_report_manifest"}
+
+    reopened = _run(workspace, "open")
+    assert reopened.reused is True
+    intent = parse_intent_from_query("打开基因家族结果报告")
+    assert intent["stages"] == ["gene_family_report"]
+    assert intent["params"]["reportAction"] == "open"
+
+    print(json.dumps({
+        "status": "passed",
+        "mode": args.mode,
+        "sequence_count": count,
+        "report": str(generated.report_path),
+    }, ensure_ascii=False))
+    return 0
+
+
+if __name__ == "__main__":
+    raise SystemExit(main())