debase 0.5.1__py3-none-any.whl → 0.6.0__py3-none-any.whl

debase/_version.py

@@ -1,3 +1,3 @@
 """Version information."""
 
-__version__ = "0.5.1"
+__version__ = "0.6.0"

debase/lineage_format.py

@@ -35,7 +35,6 @@ import logging
 import os
 import pickle
 import re
-import sqlite3
 import sys
 import time
 from concurrent.futures import ThreadPoolExecutor, as_completed
@@ -137,8 +136,7 @@ SUBSTRATE_CACHE_FILE: Path = CACHE_DIR / "substrate_smiles_cache.pkl"
 CANONICAL_CACHE_FILE: Path = CACHE_DIR / "canonical_smiles_cache.pkl"
 CACHE_DIR.mkdir(parents=True, exist_ok=True)
 
-# Local PubChem DB (optional) --------------------------------------------------------
-PUBCHEM_DB_PATH: Path = Path(__file__).parent.parent.parent / "data" / "iupac2smiles.db"
+# API endpoints for IUPAC to SMILES conversion --------------------------------------
 
 # Gemini API configuration -----------------------------------------------------------
 GEMINI_API_KEY: str = os.environ.get("GEMINI_API_KEY", "")
@@ -323,37 +321,7 @@ SUBSTRATE_CACHE: Dict[str, str] = _load_pickle(SUBSTRATE_CACHE_FILE)
 CANONICAL_CACHE: Dict[str, str] = _load_pickle(CANONICAL_CACHE_FILE)
 
 
-# --- Database lookup ---------------------------------------------------------------
-class PubChemDB:
-    """Very thin wrapper around a local SQLite mapping IUPAC -> SMILES."""
-
-    def __init__(self, path: Path | str) -> None:
-        self.path = Path(path)
-        self._conn: Optional[sqlite3.Connection] = None
-        if not self.path.exists():
-            log.warning("Local PubChem DB not found at %s", self.path)
-
-    def _connect(self) -> sqlite3.Connection:
-        if self._conn is None:
-            self._conn = sqlite3.connect(str(self.path))
-        return self._conn
-
-    def lookup(self, name: str) -> Optional[str]:
-        if not self.path.exists():
-            return None
-        sql = "SELECT smiles FROM x WHERE name = ? LIMIT 1"
-        try:
-            # Create a new connection for thread safety
-            conn = sqlite3.connect(str(self.path))
-            cur = conn.execute(sql, (name.lower(),))
-            row = cur.fetchone()
-            conn.close()
-            return row[0] if row else None
-        except Exception:  # pragma: no cover
-            return None
-
-
-PC_DB = PubChemDB(PUBCHEM_DB_PATH)
+# --- Removed local database - using only online APIs -------------------------------
 
 
 # === 5. SEQUENCE / MUTATION HELPERS ================================================
@@ -481,12 +449,7 @@ def _name_to_smiles(name: str, is_substrate: bool) -> str:
     if not name or name.lower() in ['nan', 'none', 'null', 'n/a', 'na', '']:
         return ""
     
-    # 1. Local DB (fast, offline)
-    db_smiles = PC_DB.lookup(name)
-    if db_smiles:
-        return db_smiles
-
-    # 2. OPSIN (if installed) ---------------------------------------------------
+    # 1. OPSIN (if installed) - fast and reliable for IUPAC names
     try:
         import subprocess
 
@@ -503,12 +466,7 @@ def _name_to_smiles(name: str, is_substrate: bool) -> str:
     except FileNotFoundError:
         pass  # OPSIN not installed
 
-    # 3. Gemini search (for complex compounds) ---------------------------------
-    gemini_smiles = search_smiles_with_gemini(name)
-    if gemini_smiles:
-        return gemini_smiles
-    
-    # 4. PubChem PUG REST (online) ---------------------------------------------
+    # 2. PubChem PUG REST API (online) - comprehensive database
     try:
         import requests
 
@@ -521,6 +479,11 @@ def _name_to_smiles(name: str, is_substrate: bool) -> str:
             return pug_smiles
     except Exception:  # pragma: no cover
         pass
+    
+    # 3. Gemini search (for complex compounds) - AI fallback
+    gemini_smiles = search_smiles_with_gemini(name)
+    if gemini_smiles:
+        return gemini_smiles
 
     # Return empty string if all methods fail
     return ""

{debase-0.5.1.dist-info → debase-0.6.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: debase
-Version: 0.5.1
+Version: 0.6.0
 Summary: Enzyme lineage analysis and sequence extraction package
 Home-page: https://github.com/YuemingLong/DEBase
 Author: DEBase Team

{debase-0.5.1.dist-info → debase-0.6.0.dist-info}/RECORD

@@ -1,16 +1,16 @@
 debase/__init__.py,sha256=YeKveGj_8fwuu5ozoK2mUU86so_FjiCwsvg1d_lYVZU,586
 debase/__main__.py,sha256=LbxYt2x9TG5Ced7LpzzX_8gkWyXeZSlVHzqHfqAiPwQ,160
-debase/_version.py,sha256=iDuv12GVbaAFXFufv3yqwn-3Hwv9Kua4nJZQ-gUNJXw,49
+debase/_version.py,sha256=C3pfAAiv8eypvKmkKXrxXIwx9sLe4TtShkzld5GJJ7k,49
 debase/build_db.py,sha256=bW574GxsL1BJtDwM19urLbciPcejLzfraXZPpzm09FQ,7167
 debase/cleanup_sequence.py,sha256=qKAou871Eri4SDQMz-XCfD3D2BuuINxSxzJZMACJ7p4,73313
 debase/enzyme_lineage_extractor.py,sha256=C2rVFyM84TvDy7hvk_xIeVSdh1F6WSe4QQB8B8QrPC4,168026
-debase/lineage_format.py,sha256=Omb3oug0oEfQLcC_5XsbACvTDV7PFIIlGRtOhxC7Nwo,57844
+debase/lineage_format.py,sha256=BE8uW1XUCmxlcYKiD7QveF4r99xObfGf1vP1rZzJTV8,56525
 debase/reaction_info_extractor.py,sha256=9QXbtp0RSP6QMqQ_azBWDceGIqiw2JPCg3eJ0Ba_lxA,167849
 debase/substrate_scope_extractor.py,sha256=ydU6iZVRw3fLyQ8kIQs6ZuruBMvM4mMXIeGuPgCUOn4,115956
 debase/wrapper.py,sha256=0z1BRvs3pzuPV_sgJxrBVmX_IXqwX3tB4u0GXdSgR3c,24568
-debase-0.5.1.dist-info/licenses/LICENSE,sha256=5sk9_tcNmr1r2iMIUAiioBo7wo38u8BrPlO7f0seqgE,1075
-debase-0.5.1.dist-info/METADATA,sha256=GoaBFl0kdh8dtrApBTMoLWH6fe5GYLiSYC5JrohbPcI,4047
-debase-0.5.1.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-debase-0.5.1.dist-info/entry_points.txt,sha256=hUcxA1b4xORu-HHBFTe9u2KTdbxPzt0dwz95_6JNe9M,48
-debase-0.5.1.dist-info/top_level.txt,sha256=2BUeq-4kmQr0Rhl06AnRzmmZNs8WzBRK9OcJehkcdk8,7
-debase-0.5.1.dist-info/RECORD,,
+debase-0.6.0.dist-info/licenses/LICENSE,sha256=5sk9_tcNmr1r2iMIUAiioBo7wo38u8BrPlO7f0seqgE,1075
+debase-0.6.0.dist-info/METADATA,sha256=o_Iq-jjk_ekBLLheIQZuLEeC7tWbduq70b5jqfiCmro,4047
+debase-0.6.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+debase-0.6.0.dist-info/entry_points.txt,sha256=hUcxA1b4xORu-HHBFTe9u2KTdbxPzt0dwz95_6JNe9M,48
+debase-0.6.0.dist-info/top_level.txt,sha256=2BUeq-4kmQr0Rhl06AnRzmmZNs8WzBRK9OcJehkcdk8,7
+debase-0.6.0.dist-info/RECORD,,