napistu 0.3.6__py3-none-any.whl → 0.4.0__py3-none-any.whl

napistu/network/net_propagation.py

@@ -5,7 +5,7 @@ import pandas as pd
 import numpy as np
 import igraph as ig
 
-from napistu.network.napistu_graph_core import NapistuGraph
+from napistu.network.ng_core import NapistuGraph
 
 
 def personalized_pagerank_by_attribute(

napistu/network/{napistu_graph_core.py → ng_core.py}

@@ -36,7 +36,7 @@ class NapistuGraph(ig.Graph):
     ----------
     is_reversed : bool
         Whether the graph edges have been reversed from their original direction
-    graph_type : str or None
+    wiring_approach : str or None
         Type of graph (e.g., 'bipartite', 'regulatory', 'surrogate')
     weighting_strategy : str or None
         Strategy used for edge weighting (e.g., 'topology', 'mixed', 'calibrated')
@@ -101,7 +101,7 @@ class NapistuGraph(ig.Graph):
         # Initialize metadata
         self._metadata = {
             "is_reversed": False,
-            "graph_type": None,
+            "wiring_approach": None,
             "weighting_strategy": None,
             "creation_params": {},
         }
@@ -147,6 +147,21 @@ class NapistuGraph(ig.Graph):
 
         return new_graph
 
+    @property
+    def is_reversed(self) -> bool:
+        """Check if the graph has been reversed."""
+        return self._metadata["is_reversed"]
+
+    @property
+    def wiring_approach(self) -> Optional[str]:
+        """Get the graph type (bipartite, regulatory, etc.)."""
+        return self._metadata["wiring_approach"]
+
+    @property
+    def weighting_strategy(self) -> Optional[str]:
+        """Get the weighting strategy used."""
+        return self._metadata["weighting_strategy"]
+
     def reverse_edges(self) -> None:
         """
         Reverse all edges in the graph.
@@ -181,20 +196,47 @@ class NapistuGraph(ig.Graph):
 
         return None
 
-    @property
-    def is_reversed(self) -> bool:
-        """Check if the graph has been reversed."""
-        return self._metadata["is_reversed"]
+    def remove_isolated_vertices(self):
+        """
+        Remove vertices that have no edges (degree 0) from the graph.
 
-    @property
-    def graph_type(self) -> Optional[str]:
-        """Get the graph type (bipartite, regulatory, etc.)."""
-        return self._metadata["graph_type"]
 
-    @property
-    def weighting_strategy(self) -> Optional[str]:
-        """Get the weighting strategy used."""
-        return self._metadata["weighting_strategy"]
+        Returns
+        -------
+        None
+            The graph is modified in-place.
+
+        """
+
+        # Find isolated vertices (degree 0)
+        isolated_vertices = self.vs.select(_degree=0)
+
+        if len(isolated_vertices) == 0:
+            logger.info("No isolated vertices found to remove")
+            return
+
+        # Get vertex names/indices for logging (up to 5 examples)
+        vertex_names = []
+        for v in isolated_vertices[:5]:
+            # Use vertex name if available, otherwise use index
+            name = (
+                v["name"]
+                if "name" in v.attributes() and v["name"] is not None
+                else str(v.index)
+            )
+            vertex_names.append(name)
+
+        # Create log message
+        examples_str = ", ".join(f"'{name}'" for name in vertex_names)
+        if len(isolated_vertices) > 5:
+            examples_str += f" (and {len(isolated_vertices) - 5} more)"
+
+        logger.info(
+            f"Removed {len(isolated_vertices)} isolated vertices: [{examples_str}]"
+        )
+
+        # Remove the isolated vertices
+        self.delete_vertices(isolated_vertices)
 
     def set_metadata(self, **kwargs) -> None:
         """
@@ -252,7 +294,7 @@ class NapistuGraph(ig.Graph):
         base_str = super().__str__()
         metadata_str = (
             f"Reversed: {self.is_reversed}, "
-            f"Type: {self.graph_type}, "
+            f"Type: {self.wiring_approach}, "
             f"Weighting: {self.weighting_strategy}"
         )
         return f"{base_str}\nNapistuGraph metadata: {metadata_str}"

napistu/network/ng_utils.py

@@ -17,12 +17,12 @@ import pandas as pd
 from napistu import sbml_dfs_core
 from napistu import source
 from napistu.network import net_create
-from napistu.network.napistu_graph_core import NapistuGraph
+from napistu.network.ng_core import NapistuGraph
 
 from napistu.constants import SBML_DFS
 from napistu.constants import SOURCE_SPEC
 from napistu.identifiers import _validate_assets_sbml_ids
-from napistu.network.constants import NAPISTU_GRAPH_TYPES
+from napistu.network.constants import GRAPH_WIRING_APPROACHES
 from napistu.network.constants import NAPISTU_GRAPH_DIRECTEDNESS
 
 logger = logging.getLogger(__name__)
@@ -138,9 +138,9 @@ def export_networks(
     model_prefix: str,
     outdir: str,
     directeds: list[bool] = [True, False],
-    graph_types: list[str] = [
-        NAPISTU_GRAPH_TYPES.BIPARTITE,
-        NAPISTU_GRAPH_TYPES.REGULATORY,
+    wiring_approaches: list[str] = [
+        GRAPH_WIRING_APPROACHES.BIPARTITE,
+        GRAPH_WIRING_APPROACHES.REGULATORY,
     ],
 ) -> None:
     """
@@ -158,10 +158,11 @@ def export_networks(
         Path to an existing directory where results should be saved
     directeds : [bool]
         List of directed types to export: a directed (True) or undirected graph be made (False)
-    graph_types : [str]
+    wiring_approaches : [str]
         Types of graphs to construct, valid values are:
             - bipartite: substrates and modifiers point to the reaction they drive, this reaction points to products
             - regulatory: non-enzymatic modifiers point to enzymes, enzymes point to substrates and products
+            - surrogate regulatory approach but with substrates upstream of enzymes
 
     Returns:
     ----------
@@ -177,24 +178,26 @@ def export_networks(
         raise FileNotFoundError(f"{outdir} does not exist")
     if not isinstance(directeds, list):
         raise TypeError(f"directeds must be a list, but was {type(directeds)}")
-    if not isinstance(graph_types, list):
-        raise TypeError(f"graph_types must be a list but was a {type(graph_types)}")
+    if not isinstance(wiring_approaches, list):
+        raise TypeError(
+            f"wiring_approaches must be a list but was a {type(wiring_approaches)}"
+        )
 
-    # iterate through provided graph_types and export each type
-    for graph_type in graph_types:
+    # iterate through provided wiring_approaches and export each type
+    for wiring_approach in wiring_approaches:
         for directed in directeds:
             export_pkl_path = _create_network_save_string(
                 model_prefix=model_prefix,
                 outdir=outdir,
                 directed=directed,
-                graph_type=graph_type,
+                wiring_approach=wiring_approach,
             )
-            print(f"Exporting {graph_type} network to {export_pkl_path}")
+            print(f"Exporting {wiring_approach} network to {export_pkl_path}")
 
             network_graph = net_create.process_napistu_graph(
                 sbml_dfs=sbml_dfs,
                 directed=directed,
-                graph_type=graph_type,
+                wiring_approach=wiring_approach,
                 verbose=True,
             )
 
@@ -206,7 +209,7 @@ def export_networks(
 def read_network_pkl(
     model_prefix: str,
     network_dir: str,
-    graph_type: str,
+    wiring_approach: str,
     directed: bool = True,
 ) -> NapistuGraph:
     """
@@ -222,10 +225,11 @@ def read_network_pkl(
         Path to a directory containing all saved networks.
     directed : bool
         Should a directed (True) or undirected graph be loaded (False)
-    graph_type : [str]
+    wiring_approach : [str]
         Type of graphs to read, valid values are:
             - bipartite: substrates and modifiers point to the reaction they drive, this reaction points to products
             - reguatory: non-enzymatic modifiers point to enzymes, enzymes point to substrates and products
+            - surrogate regulatory approach but with substrates upstream of enzymes
 
     Returns
     -------
@@ -239,15 +243,17 @@ def read_network_pkl(
         raise FileNotFoundError(f"{network_dir} does not exist")
     if not isinstance(directed, bool):
         raise TypeError(f"directed must be a bool, but was {type(directed)}")
-    if not isinstance(graph_type, str):
-        raise TypeError(f"graph_type must be a str but was a {type(graph_type)}")
+    if not isinstance(wiring_approach, str):
+        raise TypeError(
+            f"wiring_approach must be a str but was a {type(wiring_approach)}"
+        )
 
     import_pkl_path = _create_network_save_string(
-        model_prefix, network_dir, directed, graph_type
+        model_prefix, network_dir, directed, wiring_approach
     )
     if not os.path.isfile(import_pkl_path):
         raise FileNotFoundError(f"{import_pkl_path} does not exist")
-    print(f"Importing {graph_type} network from {import_pkl_path}")
+    print(f"Importing {wiring_approach} network from {import_pkl_path}")
 
     network_graph = ig.Graph.Read_Pickle(fname=import_pkl_path)
 
@@ -374,7 +380,7 @@ def read_graph_attrs_spec(graph_attrs_spec_uri: str) -> dict:
 
 # Internal utility functions
 def _create_network_save_string(
-    model_prefix: str, outdir: str, directed: bool, graph_type: str
+    model_prefix: str, outdir: str, directed: bool, wiring_approach: str
 ) -> str:
     if directed:
         directed_str = NAPISTU_GRAPH_DIRECTEDNESS.DIRECTED
@@ -382,7 +388,8 @@ def _create_network_save_string(
         directed_str = NAPISTU_GRAPH_DIRECTEDNESS.UNDIRECTED
 
     export_pkl_path = os.path.join(
-        outdir, model_prefix + "_network_" + graph_type + "_" + directed_str + ".pkl"
+        outdir,
+        model_prefix + "_network_" + wiring_approach + "_" + directed_str + ".pkl",
     )
 
     return export_pkl_path

napistu/network/paths.py

@@ -9,9 +9,9 @@ import pandas as pd
 
 from napistu import sbml_dfs_core
 from napistu import utils
-from napistu.network.napistu_graph_core import NapistuGraph
+from napistu.network.ng_core import NapistuGraph
 from napistu.network.ng_utils import get_minimal_sources_edges
-from napistu.constants import CPR_PATH_REQ_VARS
+from napistu.constants import NAPISTU_PATH_REQ_VARS
 from napistu.constants import MINI_SBO_NAME_TO_POLARITY
 from napistu.constants import MINI_SBO_TO_NAME
 from napistu.constants import SBML_DFS
@@ -391,7 +391,7 @@ def _filter_paths_by_precomputed_distances(
 ) -> pd.DataFrame:
     """Filter source -> destination pairs based on precomputed distances if they were provided."""
 
-    utils.match_pd_vars(all_species_pairs, CPR_PATH_REQ_VARS).assert_present()
+    utils.match_pd_vars(all_species_pairs, NAPISTU_PATH_REQ_VARS).assert_present()
 
     if precomputed_distances is None:
         logger.info(
@@ -402,7 +402,7 @@ def _filter_paths_by_precomputed_distances(
         if not isinstance(precomputed_distances, pd.DataFrame):
             raise TypeError('"precomputed_distances" must be a pd.DataFrame')
 
-    utils.match_pd_vars(precomputed_distances, CPR_PATH_REQ_VARS).assert_present()
+    utils.match_pd_vars(precomputed_distances, NAPISTU_PATH_REQ_VARS).assert_present()
 
     # filter to pairs which are connected in the pre-computed distances table
     valid_all_species_pairs = all_species_pairs.merge(

napistu/network/precompute.py

@@ -2,17 +2,16 @@ from __future__ import annotations
 
 import logging
 import math
-from pathlib import Path
-from typing import Union
-import io
 
 import numpy as np
 import pandas as pd
-from fs.errors import ResourceNotFound
 
-from napistu.network.napistu_graph_core import NapistuGraph
+from napistu.network.ng_core import NapistuGraph
 from napistu.network.ig_utils import validate_edge_attributes
-from napistu.utils import load_json, save_json
+from napistu.constants import NAPISTU_EDGELIST, SBML_DFS
+from napistu.network.constants import (
+    NAPISTU_GRAPH_EDGES,
+)
 
 logger = logging.getLogger(__name__)
 
@@ -22,10 +21,13 @@ def precompute_distances(
     max_steps: int = -1,
     max_score_q: float = float(1),
     partition_size: int = int(5000),
-    weights_vars: list[str] = ["weights", "upstream_weights"],
+    weights_vars: list[str] = [
+        NAPISTU_GRAPH_EDGES.WEIGHTS,
+        NAPISTU_GRAPH_EDGES.UPSTREAM_WEIGHTS,
+    ],
 ) -> pd.DataFrame:
     """
-    Pre-Compute Distances
+    Precompute Distances between all pairs of species in a NapistuGraph network.
 
     Parameters
     ----------
@@ -80,6 +82,7 @@ def precompute_distances(
     # interate through all partitions of "from" nodes and find their shortest and lowest weighted paths
     unique_partitions = vs_to_partition.index.unique().tolist()
 
+    logger.info(f"Calculating distances for {len(unique_partitions)} partitions")
     precomputed_distances = pd.concat(
         [
             _calculate_distances_subset(
@@ -93,6 +96,10 @@ def precompute_distances(
     ).query("sc_id_origin != sc_id_dest")
 
     # filter by path length and/or weight
+
+    logger.info(
+        f"Filtering distances by path length ({max_steps}) and score quantile ({max_score_q})"
+    )
     filtered_precomputed_distances = _filter_precomputed_distances(
         precomputed_distances=precomputed_distances,
         max_steps=max_steps,
@@ -103,65 +110,14 @@ def precompute_distances(
     return filtered_precomputed_distances
 
 
-def save_precomputed_distances(
-    precomputed_distances: pd.DataFrame, uri: Union[str, Path]
-) -> None:
-    """
-    Save a precomputed distances DataFrame to a JSON file.
-
-    Parameters
-    ----------
-    precomputed_distances : pd.DataFrame
-        The precomputed distances DataFrame to save
-    uri : Union[str, Path]
-        Path where to save the JSON file. Can be a local path or a GCS URI.
-
-    Raises
-    ------
-    OSError
-        If the file cannot be written to (permission issues, etc.)
-    """
-    save_json(str(uri), precomputed_distances.to_json())
-
-
-def load_precomputed_distances(uri: Union[str, Path]) -> pd.DataFrame:
-    """
-    Load a precomputed distances DataFrame from a JSON file.
-
-    Parameters
-    ----------
-    uri : Union[str, Path]
-        Path to the JSON file to load
-
-    Returns
-    -------
-    pd.DataFrame
-        The reconstructed precomputed distances DataFrame
-
-    Raises
-    ------
-    FileNotFoundError
-        If the specified file does not exist
-    """
-    try:
-        json_string = load_json(str(uri))
-        df = pd.read_json(io.StringIO(json_string))
-
-        # Convert integer columns to float
-        for col in df.columns:
-            if df[col].dtype in ["int64", "int32", "int16", "int8"]:
-                df[col] = df[col].astype(float)
-
-        return df
-    except ResourceNotFound as e:
-        raise FileNotFoundError(f"File not found: {uri}") from e
-
-
 def _calculate_distances_subset(
     napistu_graph: NapistuGraph,
     vs_to_partition: pd.DataFrame,
     one_partition: pd.DataFrame,
-    weights_vars: list[str] = ["weights", "upstream_weights"],
+    weights_vars: list[str] = [
+        NAPISTU_GRAPH_EDGES.WEIGHTS,
+        NAPISTU_GRAPH_EDGES.UPSTREAM_WEIGHTS,
+    ],
 ) -> pd.DataFrame:
     """Calculate distances from a subset of vertices to all vertices."""
 
@@ -169,14 +125,15 @@ def _calculate_distances_subset(
         pd.DataFrame(
             np.array(
                 napistu_graph.distances(
-                    source=one_partition["sc_id"], target=vs_to_partition["sc_id"]
+                    source=one_partition[SBML_DFS.SC_ID],
+                    target=vs_to_partition[SBML_DFS.SC_ID],
                 )
             ),
-            index=one_partition["sc_id"].rename("sc_id_origin"),
-            columns=vs_to_partition["sc_id"].rename("sc_id_dest"),
+            index=one_partition[SBML_DFS.SC_ID].rename(NAPISTU_EDGELIST.SC_ID_ORIGIN),
+            columns=vs_to_partition[SBML_DFS.SC_ID].rename(NAPISTU_EDGELIST.SC_ID_DEST),
         )
         .reset_index()
-        .melt("sc_id_origin", value_name="path_length")
+        .melt(NAPISTU_EDGELIST.SC_ID_ORIGIN, value_name="path_length")
         .replace([np.inf, -np.inf], np.nan, inplace=False)
         .dropna()
     )
@@ -187,16 +144,20 @@ def _calculate_distances_subset(
             pd.DataFrame(
                 np.array(
                     napistu_graph.distances(
-                        source=one_partition["sc_id"],
-                        target=vs_to_partition["sc_id"],
+                        source=one_partition[SBML_DFS.SC_ID],
+                        target=vs_to_partition[SBML_DFS.SC_ID],
                         weights=weight_type,
                     )
                 ),
-                index=one_partition["sc_id"].rename("sc_id_origin"),
-                columns=vs_to_partition["sc_id"].rename("sc_id_dest"),
+                index=one_partition[SBML_DFS.SC_ID].rename(
+                    NAPISTU_EDGELIST.SC_ID_ORIGIN
+                ),
+                columns=vs_to_partition[SBML_DFS.SC_ID].rename(
+                    NAPISTU_EDGELIST.SC_ID_DEST
+                ),
             )
             .reset_index()
-            .melt("sc_id_origin", value_name=f"path_{weight_type}")
+            .melt(NAPISTU_EDGELIST.SC_ID_ORIGIN, value_name=f"path_{weight_type}")
             .replace([np.inf, -np.inf], np.nan, inplace=False)
             .dropna()
         )
@@ -211,8 +172,8 @@ def _calculate_distances_subset(
     # note: these may be different paths! e.g., a longer path may have a lower weight than a short one
     path_summaries = d_steps.merge(
         d_weights,
-        left_on=["sc_id_origin", "sc_id_dest"],
-        right_on=["sc_id_origin", "sc_id_dest"],
+        left_on=[NAPISTU_EDGELIST.SC_ID_ORIGIN, NAPISTU_EDGELIST.SC_ID_DEST],
+        right_on=[NAPISTU_EDGELIST.SC_ID_ORIGIN, NAPISTU_EDGELIST.SC_ID_DEST],
     )
 
     # return connected species

napistu/ontologies/genodexito.py

@@ -356,7 +356,7 @@ class Genodexito:
             )
             logger.debug(
                 f"{ids.shape[0] - expanded_ids.shape[0]} "
-                "ids are not included in expanded ids"
+                "ids are not included in expanded ids. These will be filled with empty Identifiers"
             )
         else:
             matched_expanded_ids = expanded_ids
@@ -364,6 +364,10 @@ class Genodexito:
         updated_ids = ids.drop(SBML_DFS.S_IDENTIFIERS, axis=1).join(
             pd.DataFrame(matched_expanded_ids)
         )
+        # fill missing attributes with empty Identifiers
+        updated_ids[SBML_DFS.S_IDENTIFIERS] = updated_ids[
+            SBML_DFS.S_IDENTIFIERS
+        ].fillna(identifiers.Identifiers([]))
 
         setattr(sbml_dfs, "species", updated_ids)
 

napistu/ontologies/renaming.py

@@ -72,6 +72,10 @@ def rename_species_ontologies(
     updated_species = sbml_dfs.species.drop(SBML_DFS.S_IDENTIFIERS, axis=1).join(
         pd.DataFrame(species_identifiers)
     )
+    # fill missing attributes with empty Identifiers
+    updated_species[SBML_DFS.S_IDENTIFIERS] = updated_species[
+        SBML_DFS.S_IDENTIFIERS
+    ].fillna(identifiers.Identifiers([]))
 
     setattr(sbml_dfs, "species", updated_species)