synkit 1.3__tar.gz → 1.3.2__tar.gz

{synkit-1.3 → synkit-1.3.2}/.gitignore

@@ -3,15 +3,15 @@
 *.csv
 */catboost_info/*
 
-*.json
+
 *.pkl.gz
 *.pdf
-*.log
+
 .coverage
 dev/*
 *.rdf
 test_syn_reactor.py
-*.ipynb
+
 Data/Benchmark/*
 *.json.gz
 # *.png
@@ -26,5 +26,9 @@ Data/USPTO/*
 crn/*
 test_run.py
 Data/Study/CRN/case_formose/*
-Data/Study/CRN/case_glycolysis/*
-synkit/Graph/dev/*
+
+synkit/Graph/dev/*
+*.ipynb
+*.json
+*.txt
+*.log

{synkit-1.3 → synkit-1.3.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: synkit
-Version: 1.3
+Version: 1.3.2
 Summary: Utility for reaction modeling using graph grammar
 Project-URL: homepage, https://github.com/TieuLongPhan/SynKit
 Project-URL: source, https://github.com/TieuLongPhan/SynKit

{synkit-1.3 → synkit-1.3.2}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "synkit"
-version = "1.3"
+version = "1.3.2"
 description = "Utility for reaction modeling using graph grammar"
 readme = "README.md"
 long-description = { file = "CHANGELOG.md" }

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Query/kegg_extract.py

@@ -12,11 +12,15 @@ except Exception:
 
 from .kegg_api import KEGGClient
 from .kegg_parse import (
+    equation_to_text,
     get_compound_ids_from_equations,
     get_compound_ids_from_text,
     molblock_to_smiles,
     normalize_module_id,
+    orient_equation_to_module,
+    parse_equation,
     parse_kegg_field_blocks,
+    parse_module_reaction_directions,
     reaction_smiles_from_equation,
 )
 
@@ -127,7 +131,8 @@ class KEGGExtractor:
 
     def get_reaction_ids_from_module(self, module_id: str) -> list[str]:
         """
-        Collect KEGG reaction IDs from a module entry.
+        Collect KEGG reaction IDs from a module entry, preserving module order when
+        directional REACTION lines can be parsed.
 
         :param module_id:
             KEGG module identifier such as ``"M00001"``.
@@ -144,8 +149,11 @@ class KEGGExtractor:
             reaction_ids = extractor.get_reaction_ids_from_module("M00001")
         """
         text = self.client.get_text(f"get/{module_id}")
-        payloads = parse_kegg_field_blocks(text, "REACTION")
+        directions = parse_module_reaction_directions(text)
+        if directions:
+            return list(directions.keys())
 
+        payloads = parse_kegg_field_blocks(text, "REACTION")
         reaction_ids: set[str] = set()
         for payload in payloads:
             reaction_ids.update(_RID_PATTERN.findall(payload))
@@ -192,11 +200,35 @@ class KEGGExtractor:
 
             equations = extractor.get_module_equations("M00001")
         """
-        reaction_ids = self.get_reaction_ids_from_module(module_id)
-        return {
-            reaction_id: self.get_equation_for_reaction(reaction_id)
-            for reaction_id in reaction_ids
-        }
+        module_text = self.client.get_text(f"get/{module_id}")
+        directions = parse_module_reaction_directions(module_text)
+
+        reaction_ids = (
+            list(directions.keys())
+            if directions
+            else self.get_reaction_ids_from_module(module_id)
+        )
+
+        equations_by_rid: ReactionEquationMap = {}
+
+        for reaction_id in reaction_ids:
+            equation = self.get_equation_for_reaction(reaction_id)
+            if equation is None:
+                equations_by_rid[reaction_id] = None
+                continue
+
+            if reaction_id not in directions:
+                equations_by_rid[reaction_id] = equation
+                continue
+
+            left_ids, right_ids, module_arrow = directions[reaction_id]
+            parsed = parse_equation(equation)
+            oriented = orient_equation_to_module(parsed, left_ids, right_ids)
+            equations_by_rid[reaction_id] = equation_to_text(
+                oriented, arrow=module_arrow
+            )
+
+        return equations_by_rid
 
     def get_pathway_equations(
         self,

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Query/kegg_parse.py

@@ -9,6 +9,10 @@ from rdkit import Chem
 _CID_PATTERN = re.compile(r"C\d{5}")
 _MODULE_PATTERN = re.compile(r"M\d{5}")
 _TERM_PATTERN = re.compile(r"^(?:(\d+)\s+)?(C\d{5})$")
+_RID_PATTERN = re.compile(r"R\d{5}")
+_MODULE_REACTION_LINE = re.compile(
+    r"^(?P<rids>R\d{5}(?:,R\d{5})*)\s+(?P<lhs>.+?)\s*(?P<arrow><=>|->|=>|<-|<=)\s*(?P<rhs>.+)$"
+)
 
 CompoundStoich = tuple[str, int]
 CompoundRecord = Mapping[str, Any]
@@ -178,7 +182,7 @@ def parse_equation(equation: str) -> KEGGEquation:
     """
     Parse a KEGG equation string into reactants, products, and arrow type.
 
-    Supported arrows are ``<=>``, ``=>``, and ``<=``.
+    Supported arrows are ``<=>``, ``<->``, ``=>``, ``->``, ``<=``, and ``<-``.
 
     :param equation:
         KEGG equation string.
@@ -200,12 +204,21 @@ def parse_equation(equation: str) -> KEGGEquation:
     if "<=>" in equation:
         lhs, rhs = equation.split("<=>")
         reversible = True
+    elif "<->" in equation:
+        lhs, rhs = equation.split("<->")
+        reversible = True
     elif "=>" in equation:
         lhs, rhs = equation.split("=>")
         reversible = False
+    elif "->" in equation:
+        lhs, rhs = equation.split("->")
+        reversible = False
     elif "<=" in equation:
         rhs, lhs = equation.split("<=")
         reversible = False
+    elif "<-" in equation:
+        rhs, lhs = equation.split("<-")
+        reversible = False
     else:
         raise ValueError(f"Unknown KEGG equation arrow in: {equation!r}")
 
@@ -392,3 +405,96 @@ def reaction_smiles_from_equation(
             missing["products"].append(compound_id)
 
     return ".".join(reactant_smiles) + ">>" + ".".join(product_smiles), missing
+
+
+def parse_module_reaction_directions(
+    text: str,
+) -> dict[str, tuple[list[str], list[str], str]]:
+    """
+    Parse directional hints from a KEGG MODULE entry.
+
+    Returns
+    -------
+    dict
+        Mapping:
+        {
+            reaction_id: (left_compound_ids, right_compound_ids, arrow)
+        }
+    """
+    directions: dict[str, tuple[list[str], list[str], str]] = {}
+
+    in_reaction = False
+    for line in text.splitlines():
+        if line.startswith("REACTION"):
+            payload = line[len("REACTION") :].strip()  # noqa
+            in_reaction = True
+        elif in_reaction and (line.startswith(" ") or line.startswith("\t")):
+            payload = line.strip()
+        else:
+            if in_reaction:
+                break
+            continue
+
+        match = _MODULE_REACTION_LINE.match(payload)
+        if not match:
+            continue
+
+        reaction_ids = _RID_PATTERN.findall(match.group("rids"))
+        left_ids = get_compound_ids_from_text(match.group("lhs"))
+        right_ids = get_compound_ids_from_text(match.group("rhs"))
+        arrow = match.group("arrow")
+
+        for rid in reaction_ids:
+            directions[rid] = (left_ids, right_ids, arrow)
+
+    return directions
+
+
+def orient_equation_to_module(
+    parsed: KEGGEquation,
+    left_ids: list[str],
+    right_ids: list[str],
+) -> KEGGEquation:
+    """
+    Orient a parsed KEGG equation according to module direction.
+    """
+    reactant_ids = {cid for cid, _ in parsed.reactants}
+    product_ids = {cid for cid, _ in parsed.products}
+
+    left_set = set(left_ids)
+    right_set = set(right_ids)
+
+    keep_score = len(left_set & reactant_ids) + len(right_set & product_ids)
+    flip_score = len(left_set & product_ids) + len(right_set & reactant_ids)
+
+    if flip_score > keep_score:
+        return KEGGEquation(
+            reactants=list(parsed.products),
+            products=list(parsed.reactants),
+            reversible=parsed.reversible,
+        )
+
+    return parsed
+
+
+def equation_to_text(parsed: KEGGEquation, arrow: str | None = None) -> str:
+    """
+    Convert KEGGEquation back to text, optionally forcing the arrow from the
+    module hint.
+    """
+
+    def side_to_text(items: list[tuple[str, int]]) -> str:
+        parts = []
+        for cid, coeff in items:
+            parts.append(f"{coeff} {cid}" if coeff != 1 else cid)
+        return " + ".join(parts)
+
+    if arrow == "->":
+        arrow = "=>"
+    elif arrow == "<-":
+        arrow = "<="
+
+    if arrow is None:
+        arrow = "<=>" if parsed.reversible else "=>"
+
+    return f"{side_to_text(parsed.reactants)} {arrow} {side_to_text(parsed.products)}"

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/_common.py

@@ -542,7 +542,6 @@ def prepare_graph(
     source: Any,
     *,
     include_rule: bool = True,
-    integer_ids: bool = False,
     include_stoich: bool = True,
 ) -> Tuple[nx.DiGraph, str]:
     """
@@ -561,10 +560,6 @@ def prepare_graph(
         species-rule graph rather than a species-only graph.
     :type include_rule: bool
 
-    :param integer_ids:
-        Reserved compatibility flag.
-    :type integer_ids: bool
-
     :param include_stoich:
         Reserved compatibility flag.
     :type include_stoich: bool
@@ -576,7 +571,6 @@ def prepare_graph(
     :raises TypeError:
         If the input is unsupported.
     """
-    _ = integer_ids
     _ = include_stoich
 
     if isinstance(source, (nx.Graph, nx.DiGraph, nx.MultiGraph, nx.MultiDiGraph)):

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/_ir.py

@@ -88,11 +88,6 @@ class IRCanonicalEngine:
         Whether rule/reaction node semantics should be encoded into node tokens.
     :type include_rule: bool
 
-    :param integer_ids:
-        Whether integer-style IDs should be preferred when supported by the
-        upstream graph builder.
-    :type integer_ids: bool
-
     :param include_stoich:
         Whether stoichiometric edge/node information should be included in the
         canonicalization tokens.
@@ -145,7 +140,6 @@ class IRCanonicalEngine:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         wl_iters: int = 20,
         wl_digest_size: int = 16,
@@ -162,10 +156,6 @@ class IRCanonicalEngine:
             Whether rule/reaction semantics are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer IDs should be preserved or preferred when possible.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information is encoded into the graph tokens.
         :type include_stoich: bool
@@ -187,7 +177,6 @@ class IRCanonicalEngine:
         self.wl = WLCanonicalizer(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             n_iter=wl_iters,
             digest_size=wl_digest_size,

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/automorphism.py

@@ -32,11 +32,6 @@ class CRNAutomorphism:
         internal graph representation.
     :type include_rule: bool
 
-    :param integer_ids:
-        Whether integer-style identifiers should be preferred when supported by
-        the upstream graph builder.
-    :type integer_ids: bool
-
     :param include_stoich:
         Whether stoichiometric information should be included in the internal
         representation.
@@ -80,7 +75,6 @@ class CRNAutomorphism:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         wl_iters: int = 20,
         wl_digest_size: int = 16,
@@ -97,10 +91,6 @@ class CRNAutomorphism:
             Whether rule/reaction nodes are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer-style identifiers should be preferred.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information should be included.
         :type include_stoich: bool
@@ -120,7 +110,6 @@ class CRNAutomorphism:
         self.config, self.wl, self._engine = self._build_components(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,
@@ -166,7 +155,6 @@ class CRNAutomorphism:
         source: Any,
         *,
         include_rule: bool,
-        integer_ids: bool,
         include_stoich: bool,
         wl_iters: int,
         wl_digest_size: int,
@@ -183,10 +171,6 @@ class CRNAutomorphism:
             Whether rule/reaction nodes are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer-style identifiers should be preferred.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information should be included.
         :type include_stoich: bool
@@ -211,7 +195,6 @@ class CRNAutomorphism:
         wl = WLCanonicalizer(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             n_iter=wl_iters,
             digest_size=wl_digest_size,
@@ -220,7 +203,6 @@ class CRNAutomorphism:
         engine = IRCanonicalEngine(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,
@@ -234,7 +216,6 @@ class CRNAutomorphism:
         source: Any,
         *,
         include_rule: bool,
-        integer_ids: bool,
         include_stoich: bool,
         wl_iters: int,
         wl_digest_size: int,
@@ -251,10 +232,6 @@ class CRNAutomorphism:
             Whether rule/reaction nodes are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer-style identifiers should be preferred.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information should be included.
         :type include_stoich: bool
@@ -282,7 +259,6 @@ class CRNAutomorphism:
         return cls._build_fresh(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/canon.py

@@ -33,10 +33,6 @@ class CRNCanonicalizer:
         Whether rule / reaction nodes should be included explicitly in the
         canonicalization model.
     :type include_rule: bool
-    :param integer_ids:
-        Whether to normalize node identifiers into integer-based ids in the
-        internal representation.
-    :type integer_ids: bool
     :param include_stoich:
         Whether stoichiometric information should be included in the canonical
         representation and symmetry analysis.
@@ -73,7 +69,6 @@ class CRNCanonicalizer:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         wl_iters: int = 20,
         wl_digest_size: int = 16,
@@ -88,9 +83,6 @@ class CRNCanonicalizer:
         :param include_rule:
             Whether rule / reaction nodes are included in the internal model.
         :type include_rule: bool
-        :param integer_ids:
-            Whether integer ids should be used internally.
-        :type integer_ids: bool
         :param include_stoich:
             Whether stoichiometric information is included.
         :type include_stoich: bool
@@ -111,7 +103,6 @@ class CRNCanonicalizer:
         self.wl = WLCanonicalizer(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             n_iter=wl_iters,
             digest_size=wl_digest_size,
@@ -120,7 +111,6 @@ class CRNCanonicalizer:
         self._engine = IRCanonicalEngine(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/isomorphism.py

@@ -35,11 +35,6 @@ class CRNIsomorphism:
         representation.
     :type include_rule: bool
 
-    :param integer_ids:
-        Whether integer-style identifiers should be preferred when supported by
-        the upstream graph builder.
-    :type integer_ids: bool
-
     :param include_stoich:
         Whether stoichiometric information should be included in the internal
         graph representation.
@@ -80,7 +75,6 @@ class CRNIsomorphism:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         wl_iters: int = 20,
         wl_digest_size: int = 16,
@@ -97,10 +91,6 @@ class CRNIsomorphism:
             Whether rule/reaction nodes are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer-style identifiers should be preferred.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information should be included.
         :type include_stoich: bool
@@ -121,7 +111,6 @@ class CRNIsomorphism:
         self.wl = self._build_wl(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,
@@ -135,7 +124,6 @@ class CRNIsomorphism:
         source: Any,
         *,
         include_rule: bool,
-        integer_ids: bool,
         include_stoich: bool,
         wl_iters: int,
         wl_digest_size: int,
@@ -152,10 +140,6 @@ class CRNIsomorphism:
             Whether rule/reaction nodes are included.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer-style identifiers should be preferred.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric information should be included.
         :type include_stoich: bool
@@ -179,7 +163,6 @@ class CRNIsomorphism:
         return WLCanonicalizer(
             source,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             n_iter=wl_iters,
             digest_size=wl_digest_size,

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/symmetry.py

@@ -17,7 +17,6 @@ class CRNSymmetry:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         wl_iters: int = 20,
         wl_digest_size: int = 16,
@@ -26,7 +25,6 @@ class CRNSymmetry:
         self.config = config or SymmetryConfig.topological()
         self.kwargs = dict(
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             wl_iters=wl_iters,
             wl_digest_size=wl_digest_size,
@@ -37,7 +35,6 @@ class CRNSymmetry:
             n_iter=wl_iters,
             digest_size=wl_digest_size,
             include_rule=include_rule,
-            integer_ids=integer_ids,
             include_stoich=include_stoich,
             config=self.config,
         )

{synkit-1.3 → synkit-1.3.2}/synkit/CRN/Symmetry/wl_canon.py

@@ -163,11 +163,6 @@ class WLCanonicalizer:
         Whether rule nodes should be included in the prepared graph.
     :type include_rule: bool
 
-    :param integer_ids:
-        Whether integer node identifiers should be used during graph
-        preparation.
-    :type integer_ids: bool
-
     :param include_stoich:
         Whether stoichiometric edge attributes should be preserved during graph
         preparation.
@@ -226,7 +221,6 @@ class WLCanonicalizer:
         source: Any,
         *,
         include_rule: bool = True,
-        integer_ids: bool = False,
         include_stoich: bool = True,
         n_iter: int = 20,
         digest_size: int = 16,
@@ -247,11 +241,6 @@ class WLCanonicalizer:
             Whether rule nodes should be included in the prepared graph.
         :type include_rule: bool
 
-        :param integer_ids:
-            Whether integer node identifiers should be used during graph
-            preparation.
-        :type integer_ids: bool
-
         :param include_stoich:
             Whether stoichiometric edge attributes should be preserved.
         :type include_stoich: bool
@@ -301,7 +290,6 @@ class WLCanonicalizer:
         """
         self.source = source
         self.include_rule = bool(include_rule)
-        self.integer_ids = bool(integer_ids)
         self.include_stoich = bool(include_stoich)
         self.n_iter = int(n_iter)
         self.digest_size = int(digest_size)
@@ -314,7 +302,6 @@ class WLCanonicalizer:
         self._G, self._graph_type = prepare_graph(
             source,
             include_rule=self.include_rule,
-            integer_ids=self.integer_ids,
             include_stoich=self.include_stoich,
         )
 
@@ -400,7 +387,6 @@ class WLCanonicalizer:
             self.G.number_of_nodes(),
             self.G.number_of_edges(),
             self.include_rule,
-            self.integer_ids,
             self.include_stoich,
             self.n_iter,
             self.digest_size,

{synkit-1.3 → synkit-1.3.2}/synkit/Chem/Molecule/atom_features.py

@@ -192,7 +192,6 @@ class AtomFeatureExtractor:
             "partial_charge": round(float(gcharge), 3),
             "hybridization": str(atom.GetHybridization()),
             "in_ring": bool(atom.IsInRing()),
-            "implicit_hcount": int(atom.GetNumImplicitHs()),
             "neighbors": neighbor_symbols,
             "atom_map": atom_map,
         }