Files
2026-08-03 18:42:47 +08:00

249 lines
6.5 KiB
YAML
Executable File
Raw Permalink Blame History

This file contains ambiguous Unicode characters
This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.
# =============================================================================
# BotanicalBaseAgent 配置文件
# =============================================================================
# 此文件包含基因家族分析Agent的全局配置
# 修改后需要重启Agent生效
# =============================================================================
# -----------------------------------------------------------------------------
# 基础路径配置
# -----------------------------------------------------------------------------
paths:
# 工作空间根目录(分析结果存放位置)
workspace: "workspace"
# 脚本目录(Perl/R脚本位置)
scripts: "scripts"
# 数据目录
data: "data"
# Pfam HMM模型目录
pfam_hmm: "data/pfam_hmm"
# 参考序列目录(如拟南芥家族成员序列)
reference: "data/reference"
# 基因组数据目录
genomes: "data/genomes"
# 知识库文档目录
kb_docs: "kb_docs"
# 向量数据库存储目录
kb_store: "kb_store"
# -----------------------------------------------------------------------------
# 工具默认参数
# -----------------------------------------------------------------------------
tools:
# HMMER参数
hmmsearch:
# 默认使用可信阈值
use_cut_tc: true
# 备用E-value阈值(当use_cut_tc=false时使用)
evalue: 1.0e-5
# CPU核心数
cpu: 4
# 超时时间(秒)
timeout: 3600
# BLAST参数
blast:
# E-value阈值
evalue: 1.0e-5
# 序列一致性阈值(%)
identity_threshold: 30.0
# 线程数
num_threads: 4
# 输出格式
outfmt: "6 std qlen slen"
# MUSCLE参数
muscle:
# 最大迭代次数
max_iterations: 16
# 超时时间(秒)
timeout: 3600
# MAFFT参数
mafft:
# 默认算法(auto会根据序列数量自动选择)
algorithm: "auto"
# 最大迭代次数
max_iterations: 1000
# 线程数
threads: 4
# 超时时间(秒)
timeout: 7200
# 自动选择算法的阈值
auto_selection:
# 序列数<=此值时使用L-INS-i
linsi_max_seqs: 200
# 序列数<=此值时使用FFT-NS-i
fftnsi_max_seqs: 500
# trimAl参数
trimal:
# 默认修剪方法
method: "gappyout"
# 可选方法: gappyout, strict, automated1
# seqtk参数
seqtk:
# 是否优先使用seqtk(比Python实现更快)
prefer_seqtk: true
# -----------------------------------------------------------------------------
# 鉴定流程参数
# -----------------------------------------------------------------------------
identification:
# 结果合并方法
merge_method: "intersection" # intersection(交集,推荐)或 union(并集)
# 最小候选基因数警告阈值
min_candidates_warning: 5
# 最大候选基因数警告阈值(可能有假阳性)
max_candidates_warning: 500
# -----------------------------------------------------------------------------
# 蛋白特征分析参数
# -----------------------------------------------------------------------------
features:
# 理化性质计算
physicochemical:
# 是否使用R脚本(更准确)
use_r_script: true
# R脚本路径
r_script: "scripts/fasta_pI_mw.R"
# 跨膜结构域预测
transmembrane:
# 首选方法: deeptmhmm, tmhmm, simple
preferred_method: "auto"
# 疏水性扫描窗口大小
window_size: 19
# 疏水性阈值
hydrophobicity_threshold: 1.6
# 亚细胞定位预测
subcellular:
# 首选方法: local, online, feature
preferred_method: "auto"
# 本地脚本路径
local_script: "scripts/get_Plant_mPLoc_out.py"
# 在线服务超时(秒)
online_timeout: 300
# -----------------------------------------------------------------------------
# 比对和进化分析参数
# -----------------------------------------------------------------------------
alignment:
# 是否自动生成多种格式
auto_convert_formats: true
# 输出格式列表
output_formats:
- "fasta"
- "phylip"
- "nexus"
# 是否自动修剪比对
auto_trim: false
# 可视化设置
visualization:
# 是否自动生成可视化
auto_generate: false
# R脚本路径
r_script: "scripts/msa_figure.R"
# -----------------------------------------------------------------------------
# LLM配置
# -----------------------------------------------------------------------------
llm:
# 模型提供商: openai, anthropic, local
provider: "openai"
# OpenAI配置
openai:
model: "gpt-4"
temperature: 0
max_tokens: 4096
# 本地模型配置(如使用Ollama)
local:
base_url: "http://localhost:11434/v1"
model: "llama2"
temperature: 0
# -----------------------------------------------------------------------------
# 向量数据库配置
# -----------------------------------------------------------------------------
vector_store:
# 类型: faiss, chroma
type: "faiss"
# 嵌入模型
embedding:
model: "BAAI/bge-m3"
# 本地模型路径(如果已下载)
local_path: "models/bge-m3_from_web"
# 设备: cpu, cuda
device: "cpu"
# 检索参数
retrieval:
# 返回的文档数量
top_k: 5
# 相似度阈值
score_threshold: 0.5
# -----------------------------------------------------------------------------
# 搜索配置
# -----------------------------------------------------------------------------
search:
# Tavily搜索
tavily:
# 搜索深度: basic, advanced
search_depth: "advanced"
# 最大结果数
max_results: 5
# SearchAPI
searchapi:
engine: "google"
max_results: 5
# -----------------------------------------------------------------------------
# 日志配置
# -----------------------------------------------------------------------------
logging:
# 日志级别: DEBUG, INFO, WARNING, ERROR
level: "INFO"
# 日志文件
file: "logs/agent.log"
# 是否输出到控制台
console: true
# 日志格式
format: "%(asctime)s - %(name)s - %(levelname)s - %(message)s"
# -----------------------------------------------------------------------------
# 输出配置
# -----------------------------------------------------------------------------
output:
# 是否保存中间结果
save_intermediate: true
# 是否生成分析报告
generate_report: true
# 报告格式: markdown, html, pdf
report_format: "markdown"
# 结果文件命名模板
naming_template: "{species}_{family}_{date}"