Update biopython/mcp_output/mcp_plugin/mcp_service.py

biopython/mcp_output/mcp_plugin/mcp_service.py

@@ -146,7 +146,12 @@ def blast_search(payload: dict):
     """
     使用 BLAST 搜索相似序列
     Required fields: sequence
-    Optional fields: database (nt/nr), program (blastn/blastp), hitlist_size (default 20), expect (default 1e-10)
+    Optional fields: database (nt/nr/refseq_rna), program (blastn/blastp), hitlist_size (default 20), expect (default 1e-10)
+    
+    Database recommendations:
+    - refseq_rna: Faster, curated sequences (recommended for genes)
+    - nt: Comprehensive but VERY SLOW (can take 5-15 minutes)
+    - nr: For protein sequences
     """
     try:
         sequence = str(payload.get("sequence", "")).strip()
@@ -156,27 +161,45 @@ def blast_search(payload: dict):
         # Clean sequence
         sequence = "".join(sequence.split()).upper()
         
-        database = payload.get("database", "nt")
+        # 默认使用更快的 refseq_rna 数据库
+        database = payload.get("database", "refseq_rna")
         program = payload.get("program", "blastn")
         hitlist_size = int(payload.get("hitlist_size", 20))
         expect = float(payload.get("expect", 1e-10))
         
-        print(f"🔍 Submitting BLAST search: {len(sequence)} bp, database={database}, hits={hitlist_size}")
-        
-        # Submit BLAST request
-        handle = NCBIWWW.qblast(
-            program=program,
-            database=database,
-            sequence=sequence,
-            hitlist_size=hitlist_size,
-            expect=expect,
-            format_type="XML"
-        )
+        # 数据库选择建议
+        if database == "nt":
+            print("⚠️  Warning: 'nt' database is very large and may take 5-15 minutes!")
+            print("💡 Consider using 'refseq_rna' for faster results (curated sequences)")
         
-        blast_record = NCBIXML.read(handle)
-        handle.close()
-        
-        print(f"✅ BLAST search completed: {len(blast_record.alignments)} alignments found")
+        print(f"🔍 Submitting BLAST search: {len(sequence)} bp, database={database}, hits={hitlist_size}")
+        print(f"⏳ Please wait... This may take 1-15 minutes depending on database size")
+        
+        # Submit BLAST request with timeout handling
+        try:
+            handle = NCBIWWW.qblast(
+                program=program,
+                database=database,
+                sequence=sequence,
+                hitlist_size=hitlist_size,
+                expect=expect,
+                format_type="XML"
+            )
+            
+            blast_record = NCBIXML.read(handle)
+            handle.close()
+            
+            print(f"✅ BLAST search completed: {len(blast_record.alignments)} alignments found")
+        except Exception as blast_error:
+            # 如果使用 nt 失败，建议使用更快的数据库
+            if database == "nt":
+                return {
+                    "success": False,
+                    "result": None,
+                    "error": f"BLAST timeout or error with 'nt' database. Please try 'refseq_rna' instead. Original error: {str(blast_error)}"
+                }
+            else:
+                raise blast_error
         
         hits = []
         for idx, alignment in enumerate(blast_record.alignments[:hitlist_size]):
@@ -447,7 +470,12 @@ def identify_gene_species_complete(payload: dict):
     """
     完整的基因物种鉴定流程
     Required fields: sequence
-    Optional fields: min_identity (default 70), max_hits (default 20), database (default nt)
+    Optional fields: min_identity (default 70), max_hits (default 20), database (default refseq_rna)
+    
+    Database options:
+    - refseq_rna: Curated RNA sequences, FAST (1-3 min) - RECOMMENDED
+    - nt: All nucleotide sequences, SLOW (5-15 min)
+    - refseq_genomic: Genomic sequences
     
     This tool orchestrates multiple steps:
     1. Validate sequence
@@ -462,7 +490,7 @@ def identify_gene_species_complete(payload: dict):
         sequence = payload.get("sequence", "")
         min_identity = float(payload.get("min_identity", 70.0))
         max_hits = int(payload.get("max_hits", 20))
-        database = payload.get("database", "nt")
+        database = payload.get("database", "refseq_rna")  # 默认使用更快的数据库
         
         workflow_results = {
             "steps": [],
@@ -499,6 +527,10 @@ def identify_gene_species_complete(payload: dict):
         
         # Step 3: BLAST search
         print("🔍 Step 3: Running BLAST search...")
+        print(f"📊 Using database: {database}")
+        if database == "nt":
+            print("⚠️  'nt' database may take 5-15 minutes. Consider using 'refseq_rna' for faster results.")
+        
         blast_result = blast_search({
             "sequence": validated_seq,
             "database": database,