@hongmaple0820/med-scale-research-os 0.43.0

package/target-research-platform/src/analysis/stage3_deep_research/target_function.py

@@ -0,0 +1,290 @@
+"""
+靶点功能分析模块
+整合UniProt、PubMed、KEGG等数据源进行靶点功能机制分析
+"""
+from typing import Optional
+from datetime import datetime
+
+
+class TargetFunctionAnalyzer:
+    """靶点功能分析器"""
+
+    def __init__(self):
+        self.results = {}
+
+    def analyze(
+        self,
+        gene_symbol: str,
+        uniprot_data: Optional[dict] = None,
+        pubmed_data: Optional[dict] = None,
+        pathway_data: Optional[dict] = None,
+        ppi_data: Optional[dict] = None
+    ) -> dict:
+        """
+        执行靶点功能分析
+
+        Args:
+            gene_symbol: 基因符号
+            uniprot_data: UniProt数据
+            pubmed_data: PubMed文献数据
+            pathway_data: 通路数据
+            ppi_data: 蛋白互作数据
+
+        Returns:
+            功能分析结果
+        """
+        result = {
+            "success": True,
+            "gene": gene_symbol,
+            "timestamp": datetime.now().isoformat(),
+            "sections": {}
+        }
+
+        # 1. 基本信息
+        if uniprot_data:
+            result["sections"]["basic_info"] = self._extract_basic_info(uniprot_data)
+
+        # 2. 功能描述
+        if uniprot_data:
+            result["sections"]["function"] = self._extract_function(uniprot_data)
+
+        # 3. 结构域
+        if uniprot_data:
+            result["sections"]["domains"] = self._extract_domains(uniprot_data)
+
+        # 4. 通路信息
+        if pathway_data:
+            result["sections"]["pathways"] = self._extract_pathways(pathway_data)
+
+        # 5. 蛋白互作网络
+        if ppi_data:
+            result["sections"]["ppi_network"] = self._extract_ppi(ppi_data)
+
+        # 6. 文献概况
+        if pubmed_data:
+            result["sections"]["literature"] = self._extract_literature(pubmed_data)
+
+        # 7. 生成百科全书式摘要
+        result["encyclopedia"] = self._generate_encyclopedia(result["sections"])
+
+        self.results[gene_symbol] = result
+        return result
+
+    def _extract_basic_info(self, uniprot_data: dict) -> dict:
+        """提取基本信息"""
+        return {
+            "uniprot_id": uniprot_data.get("uniprot_id", ""),
+            "gene_name": uniprot_data.get("gene_name", ""),
+            "protein_name": uniprot_data.get("protein_name", ""),
+            "organism": uniprot_data.get("organism", "Homo sapiens"),
+            "sequence_length": uniprot_data.get("sequence_length", 0),
+            "molecular_weight": uniprot_data.get("molecular_weight", 0),
+            "subcellular_location": uniprot_data.get("subcellular_location", [])
+        }
+
+    def _extract_function(self, uniprot_data: dict) -> dict:
+        """提取功能描述"""
+        function_texts = uniprot_data.get("function", [])
+
+        # 合并功能描述
+        full_function = " ".join(function_texts) if function_texts else ""
+
+        return {
+            "function_description": full_function,
+            "function_summary": self._summarize_function(full_function)
+        }
+
+    def _summarize_function(self, function_text: str) -> str:
+        """生成功能摘要"""
+        if not function_text:
+            return "功能描述不可用"
+
+        # 截取前500字符作为摘要
+        if len(function_text) > 500:
+            return function_text[:500] + "..."
+        return function_text
+
+    def _extract_domains(self, uniprot_data: dict) -> list:
+        """提取结构域信息"""
+        domains = uniprot_data.get("domains", [])
+
+        return [
+            {
+                "name": domain.get("name", ""),
+                "start": domain.get("start", 0),
+                "end": domain.get("end", 0),
+                "description": domain.get("description", "")
+            }
+            for domain in domains
+        ]
+
+    def _extract_pathways(self, pathway_data: dict) -> list:
+        """提取通路信息"""
+        pathways = pathway_data.get("pathways", [])
+
+        return [
+            {
+                "name": pathway.get("name", ""),
+                "source": pathway.get("source", ""),
+                "id": pathway.get("id", ""),
+                "description": pathway.get("description", "")
+            }
+            for pathway in pathways[:20]  # 限制返回数量
+        ]
+
+    def _extract_ppi(self, ppi_data: dict) -> dict:
+        """提取蛋白互作网络"""
+        interactions = ppi_data.get("interactions", [])
+
+        # 按置信度排序
+        sorted_interactions = sorted(
+            interactions,
+            key=lambda x: x.get("score", 0),
+            reverse=True
+        )
+
+        return {
+            "total_interactions": len(interactions),
+            "top_partners": [
+                {
+                    "partner": interaction.get("partner", ""),
+                    "score": interaction.get("score", 0),
+                    "evidence": interaction.get("evidence", "")
+                }
+                for interaction in sorted_interactions[:20]
+            ]
+        }
+
+    def _extract_literature(self, pubmed_data: dict) -> dict:
+        """提取文献概况"""
+        papers = pubmed_data.get("papers", [])
+
+        if not papers:
+            return {
+                "total_papers": 0,
+                "recent_papers": [],
+                "research_trends": []
+            }
+
+        # 按年份统计
+        year_counts = {}
+        for paper in papers:
+            year = paper.get("year", "")
+            if year:
+                year_counts[year] = year_counts.get(year, 0) + 1
+
+        # 最近文献
+        recent_papers = sorted(
+            papers,
+            key=lambda x: x.get("year", ""),
+            reverse=True
+        )[:10]
+
+        return {
+            "total_papers": len(papers),
+            "year_distribution": year_counts,
+            "recent_papers": [
+                {
+                    "pmid": paper.get("pmid", ""),
+                    "title": paper.get("title", ""),
+                    "year": paper.get("year", ""),
+                    "journal": paper.get("journal", "")
+                }
+                for paper in recent_papers
+            ]
+        }
+
+    def _generate_encyclopedia(self, sections: dict) -> dict:
+        """生成百科全书式摘要"""
+        basic_info = sections.get("basic_info", {})
+        function = sections.get("function", {})
+        domains = sections.get("domains", [])
+        pathways = sections.get("pathways", [])
+        ppi = sections.get("ppi_network", {})
+        literature = sections.get("literature", {})
+
+        # 生成概述
+        overview_parts = []
+
+        protein_name = basic_info.get("protein_name", "")
+        if protein_name:
+            overview_parts.append(f"{protein_name}")
+
+        gene_name = basic_info.get("gene_name", "")
+        if gene_name:
+            overview_parts.append(f"由{gene_name}基因编码")
+
+        function_desc = function.get("function_summary", "")
+        if function_desc and function_desc != "功能描述不可用":
+            overview_parts.append(function_desc)
+
+        overview = "。".join(overview_parts) + "。" if overview_parts else ""
+
+        # 关键特征
+        key_features = []
+
+        if domains:
+            domain_names = [d.get("name", "") for d in domains[:3] if d.get("name")]
+            if domain_names:
+                key_features.append(f"包含{', '.join(domain_names)}等结构域")
+
+        if pathways:
+            pathway_names = [p.get("name", "") for p in pathways[:3] if p.get("name")]
+            if pathway_names:
+                key_features.append(f"参与{', '.join(pathway_names)}等信号通路")
+
+        if ppi.get("top_partners"):
+            partners = [p.get("partner", "") for p in ppi["top_partners"][:3]]
+            if partners:
+                key_features.append(f"与{', '.join(partners)}等蛋白存在互作")
+
+        return {
+            "overview": overview,
+            "key_features": key_features,
+            "research_status": {
+                "total_publications": literature.get("total_papers", 0),
+                "research_active": literature.get("total_papers", 0) > 100
+            }
+        }
+
+    def validate_result(self, result: dict) -> tuple[bool, list[str]]:
+        """验证分析结果"""
+        issues = []
+
+        if not result.get("success"):
+            issues.append("分析失败")
+            return False, issues
+
+        sections = result.get("sections", {})
+
+        # 检查基本信息
+        if not sections.get("basic_info"):
+            issues.append("缺少基本信息")
+
+        # 检查功能描述
+        function = sections.get("function", {})
+        if not function.get("function_description"):
+            issues.append("缺少功能描述")
+
+        return len(issues) == 0, issues
+
+    def get_summary(self, gene_symbol: str) -> dict:
+        """获取分析摘要"""
+        if gene_symbol not in self.results:
+            return {"gene": gene_symbol, "status": "not_analyzed"}
+
+        result = self.results[gene_symbol]
+        sections = result.get("sections", {})
+        encyclopedia = result.get("encyclopedia", {})
+
+        return {
+            "gene": gene_symbol,
+            "status": "analyzed",
+            "has_basic_info": bool(sections.get("basic_info")),
+            "has_function": bool(sections.get("function", {}).get("function_description")),
+            "has_domains": len(sections.get("domains", [])) > 0,
+            "has_pathways": len(sections.get("pathways", [])) > 0,
+            "has_ppi": sections.get("ppi_network", {}).get("total_interactions", 0) > 0,
+            "publication_count": sections.get("literature", {}).get("total_papers", 0),
+            "overview": encyclopedia.get("overview", "")
+        }

package/target-research-platform/src/analysis/stage4_business/__init__.py

@@ -0,0 +1,12 @@
+"""
+阶段4：商业计划书与里程碑规划
+"""
+from .milestone_planner import MilestonePlanner
+from .business_plan import BusinessPlanGenerator
+from .pipeline import Stage4Pipeline
+
+__all__ = [
+    "MilestonePlanner",
+    "BusinessPlanGenerator",
+    "Stage4Pipeline"
+]