masster 0.4.0__py3-none-any.whl → 0.4.1__py3-none-any.whl

masster/data/libs/ccm.py

@@ -5,8 +5,8 @@ Workflow:
     amino acids, common organic acids, nucleotides, fatty acids, cofactors, sugars).
 - Query PubChem's PUG-REST for MolecularFormula, CanonicalSMILES and InChIKey for each name
     with retries and basic name normalization to improve matching.
-- Save results to `masster/data/examples/central_carbon_metabolites.csv`.
-- Test loading with `master.lib.Lib.import_csv`.
+- Save results to `masster/data/examples/ccm.csv`.
+- Test loading with `masster.lib.Lib.import_csv`.
 
 This is a best-effort programmatic lookup; ambiguous names may not resolve (those rows will
 have empty Formula/SMILES/InChIKey). For authoritative lists, prefer curated databases
@@ -70,6 +70,7 @@ CCM_METABOLITES = [
     "Pentose",
     "Acetaldehyde",
     "Acetic acid",
+
     # Proteinogenic amino acids (20 standard)
     "Alanine",
     "Arginine",
@@ -97,6 +98,7 @@ CCM_METABOLITES = [
     "Homocysteine",
     "S-adenosylmethionine",
     "S-adenosylhomocysteine",
+
     # Common organic acids / intermediates & related small metabolites
     "Formic acid",
     "Propionic acid",
@@ -108,6 +110,7 @@ CCM_METABOLITES = [
     "Beta-hydroxybutyrate",
     "Pyruvic acid",
     "Lactic acid",
+
     # Fatty acids (common)
     "Myristic acid",
     "Palmitic acid",
@@ -117,6 +120,7 @@ CCM_METABOLITES = [
     "Linoleic acid",
     "Alpha-linolenic acid",
     "Arachidonic acid",
+
     # Nucleobases and nucleosides
     "Adenine",
     "Guanine",
@@ -127,6 +131,7 @@ CCM_METABOLITES = [
     "Guanosine",
     "Cytidine",
     "Uridine",
+
     # Nucleotides (mono/di/tri)
     "AMP",
     "ADP",
@@ -140,6 +145,7 @@ CCM_METABOLITES = [
     "UMP",
     "UDP",
     "UTP",
+
     # Cofactors / common metabolites
     "NAD+",
     "NADH",
@@ -151,6 +157,7 @@ CCM_METABOLITES = [
     "Pantothenic acid",
     "Riboflavin",
     "Niacin",
+
     # Sugar and sugar derivatives
     "Fructose",
     "Mannose",
@@ -158,6 +165,7 @@ CCM_METABOLITES = [
     "Ribose",
     "Glucosamine",
     "N-acetylglucosamine",
+
     # Other common metabolites
     "Choline",
     "Betaine",
@@ -171,13 +179,40 @@ CCM_METABOLITES = [
 ]
 
 
+def canonicalize_smiles(smiles_str: str) -> str:
+    """
+    Canonicalize SMILES string using RDKit.
+    
+    Args:
+        smiles_str: Input SMILES string
+        
+    Returns:
+        Canonical SMILES string, or original string if canonicalization fails
+    """
+    if not smiles_str or not smiles_str.strip() or Chem is None:
+        return smiles_str
+    
+    try:
+        mol = Chem.MolFromSmiles(smiles_str, sanitize=True)
+        if mol is None:
+            return smiles_str
+        
+        # Generate canonical SMILES with isomeric information
+        canonical_smiles = Chem.MolToSmiles(mol, isomericSmiles=True, canonical=True)
+        return canonical_smiles
+        
+    except Exception:
+        # If canonicalization fails, return the original SMILES
+        return smiles_str
+
+
 def fetch_from_pubchem(name: str):
-    """Fetch formula, smiles and inchikey from PubChem by compound name.
+    """Fetch formula, smiles, inchikey, and CID from PubChem by compound name.
 
     Uses basic normalization and retries with exponential backoff. Returns
-    (formula, smiles, inchikey) or (None, None, None) on failure.
+    (formula, smiles, inchikey, cid) or (None, None, None, None) on failure.
     """
-    props = (None, None, None)
+    props = (None, None, None, None)
 
     def normalize_name(n: str) -> str:
         if not n:
@@ -219,6 +254,21 @@ def fetch_from_pubchem(name: str):
             return None
         return None
 
+    def try_query_with_cid(q: str):
+        """Query compound by name and get CID along with properties."""
+        url = f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/name/{quote(q)}/cids/JSON"
+        try:
+            r = requests.get(url, timeout=15)
+            if r.status_code == 200:
+                j = r.json()
+                if 'IdentifierList' in j and 'CID' in j['IdentifierList']:
+                    cids = j['IdentifierList']['CID']
+                    if cids:
+                        return cids[0]  # Return the first CID
+        except Exception:
+            return None
+        return None
+
     def try_query_inchikey(ik: str):
         url = (
             f"https://pubchem.ncbi.nlm.nih.gov/rest/pug/compound/inchikey/{quote(ik)}/property/"
@@ -251,8 +301,8 @@ def fetch_from_pubchem(name: str):
             r = requests.get(url, timeout=15)
             if r.status_code == 200:
                 j = r.json()
-                if "IdentifierList" in j and "CID" in j["IdentifierList"]:
-                    return j["IdentifierList"]["CID"]
+                if 'IdentifierList' in j and 'CID' in j['IdentifierList']:
+                    return j['IdentifierList']['CID']
         except Exception:
             return []
         return []
@@ -262,7 +312,13 @@ def fetch_from_pubchem(name: str):
 
     # exponential backoff attempts
     attempts = 3
+    cid = None
+    
     for i in range(attempts):
+        # First try to get the CID
+        if not cid:
+            cid = try_query_with_cid(query)
+            
         j = try_query(query)
         if j:
             try:
@@ -275,11 +331,7 @@ def fetch_from_pubchem(name: str):
                     # if SMILES missing, try a lookup by InChIKey (dedicated endpoint)
                     if not sm and ik:
                         j2 = try_query_inchikey(ik)
-                        if (
-                            j2
-                            and "PropertyTable" in j2
-                            and "Properties" in j2["PropertyTable"]
-                        ):
+                        if j2 and "PropertyTable" in j2 and "Properties" in j2["PropertyTable"]:
                             p2 = j2["PropertyTable"]["Properties"][0]
                             sm = p2.get("CanonicalSMILES") or sm
                             inchi = inchi or p2.get("InChI")
@@ -296,38 +348,39 @@ def fetch_from_pubchem(name: str):
                     # if still no SMILES, try fetching CIDs from InChIKey and query a CID record
                     if not sm and ik:
                         cids = try_get_cids_from_inchikey(ik)
-                        for cid in (cids or [])[:5]:
-                            j3 = try_query_cid(cid)
-                            if (
-                                j3
-                                and "PropertyTable" in j3
-                                and "Properties" in j3["PropertyTable"]
-                            ):
+                        for cid_from_ik in (cids or [])[:5]:
+                            if not cid:  # Only set CID if we don't have one yet
+                                cid = cid_from_ik
+                            j3 = try_query_cid(cid_from_ik)
+                            if j3 and "PropertyTable" in j3 and "Properties" in j3["PropertyTable"]:
                                 p3 = j3["PropertyTable"]["Properties"][0]
                                 sm = p3.get("CanonicalSMILES") or sm
                                 if sm:
                                     break
 
-                    return (mf, sm, ik)
+                    return (mf, sm, ik, cid)
             except Exception:
                 pass
-        time.sleep(1 + 2**i)
+        time.sleep(1 + 2 ** i)
 
     # final fallback: try raw name without normalization
+    if not cid:
+        cid = try_query_with_cid(name)
+        
     j = try_query(name)
     if j and "PropertyTable" in j and "Properties" in j["PropertyTable"]:
         p = j["PropertyTable"]["Properties"][0]
-        return (p.get("MolecularFormula"), p.get("CanonicalSMILES"), p.get("InChIKey"))
+        return (p.get("MolecularFormula"), p.get("CanonicalSMILES"), p.get("InChIKey"), cid)
 
     return props
 
 
-def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
+def generate_csv(out_path: str = "ccm.csv"):
     rows = []
     for name in CCM_METABOLITES:
-        formula, smiles, inchikey = (None, None, None)
+        formula, smiles, inchikey, cid = (None, None, None, None)
         if requests is not None:
-            formula, smiles, inchikey = fetch_from_pubchem(name)
+            formula, smiles, inchikey, cid = fetch_from_pubchem(name)
 
         # Neutralize charged molecular formulas (e.g., trailing +, -, 2+, 3-)
         # by adjusting the hydrogen count accordingly and removing the explicit charge.
@@ -336,15 +389,13 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
                 return fmt
             s = fmt.strip()
             # normalize common unicode superscripts (²³¹⁺⁻) to ascii
-            sup_map = str.maketrans(
-                {
-                    "²": "2",
-                    "³": "3",
-                    "¹": "1",
-                    "⁺": "+",
-                    "⁻": "-",
-                },
-            )
+            sup_map = str.maketrans({
+                "²": "2",
+                "³": "3",
+                "¹": "1",
+                "⁺": "+",
+                "⁻": "-",
+            })
             s = s.translate(sup_map)
             # Remove enclosing brackets if present, e.g. [C6H5O7]2-
             if s.startswith("[") and s.endswith("]"):
@@ -361,17 +412,13 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
             # determine magnitude and sign for patterns like '2-' or '-2' or '+2' or '3+'
             sign = 1
             mag = 1
-            if charge_str[0] in "+-":
+            if charge_str[0] in '+-':
                 # formats like '-2' or '+2' or '-' or '+'
-                sign = -1 if charge_str[0] == "-" else 1
-                mag = (
-                    int(charge_str[1:])
-                    if len(charge_str) > 1 and charge_str[1:].isdigit()
-                    else 1
-                )
-            elif charge_str[-1] in "+-":
+                sign = -1 if charge_str[0] == '-' else 1
+                mag = int(charge_str[1:]) if len(charge_str) > 1 and charge_str[1:].isdigit() else 1
+            elif charge_str[-1] in '+-':
                 # formats like '2-' or '3+'
-                sign = -1 if charge_str[-1] == "-" else 1
+                sign = -1 if charge_str[-1] == '-' else 1
                 mag = int(charge_str[:-1]) if charge_str[:-1].isdigit() else 1
 
             # parse element counts from base formula
@@ -414,7 +461,7 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
             for el in elems:
                 if el in counts:
                     n = counts[el]
-                    parts.append(f"{el}{n if n != 1 else ''}")
+                    parts.append(f"{el}{n if n!=1 else ''}")
             new_formula = "".join(parts)
             return new_formula
 
@@ -443,11 +490,7 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
                     q = a.GetFormalCharge()
                     if q > 0:
                         # remove up to q hydrogen neighbors (by index)
-                        h_neighbors = [
-                            nbr.GetIdx()
-                            for nbr in a.GetNeighbors()
-                            if nbr.GetSymbol() == "H"
-                        ]
+                        h_neighbors = [nbr.GetIdx() for nbr in a.GetNeighbors() if nbr.GetSymbol() == "H"]
                         remove = h_neighbors[: min(len(h_neighbors), q)]
                         to_remove.extend(remove)
                     elif q < 0:
@@ -486,21 +529,32 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
             smiles = neutralize_smiles(smiles) if smiles else smiles
         except Exception:
             pass
-        rows.append(
-            {
-                "Name": name,
-                "Formula": formula or "",
-                "SMILES": smiles or "",
-                "InChIKey": inchikey or "",
-            },
-        )
+        
+        # Canonicalize SMILES after neutralization
+        try:
+            smiles = canonicalize_smiles(smiles) if smiles else smiles
+        except Exception:
+            pass
+            
+        # Format the database ID and database name
+        db_id = f"CID:{cid}" if cid else ""
+        db = "pubchem" if cid else ""
+        
+        rows.append({
+            "Name": name,
+            "Formula": formula or "",
+            "SMILES": smiles or "",
+            "InChIKey": inchikey or "",
+            "db_id": db_id,
+            "db": db,
+        })
 
     # Ensure output directory exists (data/libs)
-    out_dir = os.path.join("master", "data", "libs")
+    out_dir = os.path.join('masster', 'data', 'libs')
     os.makedirs(out_dir, exist_ok=True)
     out_path_full = os.path.join(out_dir, os.path.basename(out_path))
 
-    fieldnames = ["Name", "Formula", "SMILES", "InChIKey"]
+    fieldnames = ["Name", "Formula", "SMILES", "InChIKey", "db_id", "db"]
     with open(out_path_full, "w", newline="", encoding="utf-8") as f:
         writer = csv.DictWriter(f, fieldnames=fieldnames)
         writer.writeheader()
@@ -512,11 +566,11 @@ def generate_csv(out_path: str = "central_carbon_metabolites.csv"):
 
 
 def test_load_with_lib(csv_path: str):
-    """Try to load the generated CSV using master.lib.Lib.import_csv."""
+    """Try to load the generated CSV using masster.lib.Lib.import_csv."""
     try:
-        from master.lib import Lib
+        from masster.lib import Lib
     except Exception as e:
-        print(f"Cannot import master.lib.Lib: {e}")
+        print(f"Cannot import masster.lib.Lib: {e}")
         return False
 
     try:

masster/data/libs/central_carbon_README.md

@@ -1,6 +1,6 @@
 central_carbon_metabolites.csv
 
-This folder contains example compound lists used by the master package.
+This folder contains example compound lists used by the masster package.
 
 Files:
 - central_carbon_metabolites.csv: a best-effort list of central carbon metabolism related