napistu 0.4.2__py3-none-any.whl → 0.4.4__py3-none-any.whl

napistu/source.py

@@ -1,8 +1,14 @@
 from __future__ import annotations
 
+import numpy as np
 import pandas as pd
+from typing import Optional
+
 from napistu import indices
-from napistu.constants import SOURCE_SPEC
+from napistu import sbml_dfs_core
+from napistu import sbml_dfs_utils
+from napistu.statistics import hypothesis_testing
+from napistu.constants import SBML_DFS_SCHEMA, SCHEMA_DEFS, SOURCE_SPEC
 
 
 class Source:
@@ -41,11 +47,18 @@ class Source:
             Creates an empty source object. This is typically used when creating an SBML_dfs
             object from a single source.
         pw_index : indices.PWIndex
+            a pathway index object containing the pathway_id and other metadata
 
         Returns
         -------
         None.
 
+        Raises
+        ------
+        ValueError:
+            if pw_index is not a indices.PWIndex
+        ValueError:
+            if SOURCE_SPEC.MODEL is not present in source_df
         """
 
         if init is True:
@@ -101,8 +114,27 @@ def create_source_table(
     """
     Create Source Table
 
-    Create a table with one row per "new_id" and a Source object created from the union
-      of "old_id" Source objects
+    Create a table with one row per "new_id" and a Source object created from the unionof "old_id" Source objects
+
+    Parameters
+    ----------
+    lookup_table: pd.Series
+        a pd.Series containing the index of the table to create a source table for
+    table_schema: dict
+        a dictionary containing the schema of the table to create a source table for
+    pw_index: indices.PWIndex
+        a pathway index object containing the pathway_id and other metadata
+
+    Returns
+    -------
+    source_table: pd.DataFrame
+        a pd.DataFrame containing the index of the table to create a source table for
+        with one row per "new_id" and a Source object created from the union of "old_id" Source objects
+
+    Raises
+    ------
+    ValueError:
+        if SOURCE_SPEC.SOURCE is not present in table_schema
     """
 
     if SOURCE_SPEC.SOURCE not in table_schema.keys():
@@ -142,8 +174,27 @@ def merge_sources(source_list: list | pd.Series) -> Source:
 
     Merge a list of Source objects into a single Source object
 
+    Parameters
+    ----------
+    source_list: list | pd.Series
+        a list of Source objects or a pd.Series of Source objects
+
+    Returns
+    -------
+    source: Source
+        a Source object created from the union of the Source objects in source_list
+
+    Raises
+    ------
+    TypeError:
+        if source_list is not a list or pd.Series
     """
 
+    if not isinstance(source_list, (list, pd.Series)):
+        raise TypeError(
+            f"source_list must be a list or pd.Series, but was a {type(source_list).__name__}"
+        )
+
     # filter to non-empty sources
     # empty sources have only been initialized; a merge hasn't occured
     existing_sources = [s.source is not None for s in source_list]
@@ -160,28 +211,35 @@ def merge_sources(source_list: list | pd.Series) -> Source:
     return Source(pd.concat(existing_source_list))
 
 
-def unnest_sources(
-    source_table: pd.DataFrame, source_var: str, verbose: bool = False
-) -> pd.DataFrame:
+def unnest_sources(source_table: pd.DataFrame, verbose: bool = False) -> pd.DataFrame:
     """
     Unnest Sources
 
     Take a pd.DataFrame containing an array of Sources and
     return one-row per source.
 
-    Parameters:
+    Parameters
+    ----------
     source_table: pd.DataFrame
         a table containing an array of Sources
-    source_var: str
-        variable containing Sources
+    verbose: bool
+        print progress
 
-    Returns:
+    Returns
+    -------
     pd.Dataframe containing the index of source_table but expanded
     to include one row per source
 
     """
 
     sources = list()
+
+    table_type = sbml_dfs_utils.infer_entity_type(source_table)
+    source_table_schema = SBML_DFS_SCHEMA.SCHEMA[table_type]
+    if SCHEMA_DEFS.SOURCE not in source_table_schema.keys():
+        raise ValueError(f"{table_type} does not have a source attribute")
+
+    source_var = source_table_schema[SCHEMA_DEFS.SOURCE]
     source_table_index = source_table.index.to_frame().reset_index(drop=True)
 
     for i in range(source_table.shape[0]):
@@ -216,53 +274,73 @@ def unnest_sources(
     return pd.concat(sources)
 
 
-def greedy_set_coverge_of_sources(
-    source_df: pd.DataFrame, table_schema: dict
+def source_set_coverage(
+    select_sources_df: pd.DataFrame,
+    source_total_counts: Optional[pd.Series] = None,
+    sbml_dfs: Optional[sbml_dfs_core.SBML_dfs] = None,
 ) -> pd.DataFrame:
     """
     Greedy Set Coverage of Sources
 
-    Apply the greedy set coverge algorithm to find the minimal set of
-    sources which cover all entries
+    Find the set of pathways covering `select_sources_df`. If `all_sources_df`
+    is provided pathways will be selected iteratively based on statistical
+    enrichment. If `all_sources_df` is not provided, the largest pathways
+    will be chosen iteratively.
 
-    Parameters:
-    source_df: pd.DataFrame
+    Parameters
+    ----------
+    select_sources_df: pd.DataFrame
         pd.Dataframe containing the index of source_table but expanded to
         include one row per source. As produced by source.unnest_sources()
-
-    Returns:
+    source_total_counts: pd.Series
+        pd.Series containing the total counts of each source. As produced by
+        source.get_source_total_counts()
+    sbml_dfs: sbml_dfs_core.SBML_dfs
+        if `source_total_counts` is provided then `sbml_dfs` must be provided
+        to calculate the total number of entities in the table.
+
+    Returns
+    -------
     minimial_sources: [str]
         A list of pathway_ids of the minimal source set
 
     """
 
+    table_type = sbml_dfs_utils.infer_entity_type(select_sources_df)
+    pk = SBML_DFS_SCHEMA.SCHEMA[table_type][SCHEMA_DEFS.PK]
+
+    if source_total_counts is not None:
+        if sbml_dfs is None:
+            raise ValueError(
+                "If `source_total_counts` is provided, `sbml_dfs` must be provided to calculate the total number of entities in the table."
+            )
+        n_total_entities = sbml_dfs.get_table(table_type).shape[0]
+
     # rollup pathways with identical membership
-    deduplicated_sources = _deduplicate_source_df(source_df, table_schema)
+    deduplicated_sources = _deduplicate_source_df(select_sources_df)
 
     unaccounted_for_members = deduplicated_sources
     retained_pathway_ids = []
-
     while unaccounted_for_members.shape[0] != 0:
         # find the pathway with the most members
-        pathway_members = unaccounted_for_members.groupby(SOURCE_SPEC.PATHWAY_ID).size()
-        top_pathway = pathway_members[pathway_members == max(pathway_members)].index[0]
+
+        if source_total_counts is None:
+            top_pathway = _select_top_pathway_by_size(unaccounted_for_members)
+        else:
+            top_pathway = _select_top_pathway_by_enrichment(
+                unaccounted_for_members, source_total_counts, n_total_entities, pk
+            )
+
+        if top_pathway is None:
+            break
+
         retained_pathway_ids.append(top_pathway)
 
         # remove all members associated with the top pathway
-        members_captured = (
-            unaccounted_for_members[
-                unaccounted_for_members[SOURCE_SPEC.PATHWAY_ID] == top_pathway
-            ]
-            .index.get_level_values(table_schema["pk"])
-            .tolist()
+        unaccounted_for_members = _update_unaccounted_for_members(
+            top_pathway, unaccounted_for_members
         )
 
-        unaccounted_for_members = unaccounted_for_members[
-            ~unaccounted_for_members.index.get_level_values(table_schema["pk"]).isin(
-                members_captured
-            )
-        ]
-
     minimial_sources = deduplicated_sources[
         deduplicated_sources[SOURCE_SPEC.PATHWAY_ID].isin(retained_pathway_ids)
     ].sort_index()
@@ -270,9 +348,39 @@ def greedy_set_coverge_of_sources(
     return minimial_sources
 
 
-def _deduplicate_source_df(source_df: pd.DataFrame, table_schema: dict) -> pd.DataFrame:
+def get_source_total_counts(
+    sbml_dfs: sbml_dfs_core.SBML_dfs, entity_type: str
+) -> pd.Series:
+    """
+    Get the total counts of each source.
+
+    Parameters
+    ----------
+    sbml_dfs: sbml_dfs_core.SBML_dfs
+        sbml_dfs object containing the table to get the total counts of
+    entity_type: str
+        the type of entity to get the total counts of
+
+    Returns
+    -------
+    source_total_counts: pd.Series
+        pd.Series containing the total counts of each source.
+    """
+
+    all_sources_table = unnest_sources(sbml_dfs.get_table(entity_type))
+    source_total_counts = all_sources_table.value_counts(SOURCE_SPEC.PATHWAY_ID).rename(
+        "total_counts"
+    )
+
+    return source_total_counts
+
+
+def _deduplicate_source_df(source_df: pd.DataFrame) -> pd.DataFrame:
     """Combine entries in a source table when multiple models have the same members."""
 
+    table_type = sbml_dfs_utils.infer_entity_type(source_df)
+    source_table_schema = SBML_DFS_SCHEMA.SCHEMA[table_type]
+
     # drop entries which are missing required attributes and throw an error if none are left
     REQUIRED_NON_NA_ATTRIBUTES = [SOURCE_SPEC.PATHWAY_ID]
     indexed_sources = (
@@ -296,7 +404,11 @@ def _deduplicate_source_df(source_df: pd.DataFrame, table_schema: dict) -> pd.Da
                 {
                     SOURCE_SPEC.PATHWAY_ID: p,
                     "membership_string": "_".join(
-                        set(indexed_sources.loc[[p]][table_schema["pk"]].tolist())
+                        set(
+                            indexed_sources.loc[[p]][
+                                source_table_schema[SCHEMA_DEFS.PK]
+                            ].tolist()
+                        )
                     ),
                 }
                 for p in pathways
@@ -320,16 +432,16 @@ def _deduplicate_source_df(source_df: pd.DataFrame, table_schema: dict) -> pd.Da
 
     merged_sources = pd.concat(
         [
-            _collapse_by_membership_string(s, membership_categories, table_schema)  # type: ignore
+            _collapse_by_membership_string(s, membership_categories, source_table_schema)  # type: ignore
             for s in category_index.tolist()
         ]
     )
     merged_sources[SOURCE_SPEC.INDEX_NAME] = merged_sources.groupby(
-        table_schema["pk"]
+        source_table_schema[SCHEMA_DEFS.PK]
     ).cumcount()
 
     return merged_sources.set_index(
-        [table_schema["pk"], SOURCE_SPEC.INDEX_NAME]
+        [source_table_schema[SCHEMA_DEFS.PK], SOURCE_SPEC.INDEX_NAME]
     ).sort_index()
 
 
@@ -345,7 +457,10 @@ def _collapse_by_membership_string(
     return pd.DataFrame(
         [
             pd.concat(
-                [pd.Series({table_schema["pk"]: ms}), collapsed_source_membership]
+                [
+                    pd.Series({table_schema[SCHEMA_DEFS.PK]: ms}),
+                    collapsed_source_membership,
+                ]
             )
             for ms in membership_string.split("_")
         ]
@@ -398,3 +513,91 @@ def _safe_source_merge(member_Sources: Source | list) -> Source:
         return merge_sources(member_Sources.tolist())
     else:
         raise TypeError("Expecting source.Source or pd.Series")
+
+
+def _select_top_pathway_by_size(unaccounted_for_members: pd.DataFrame) -> str:
+
+    pathway_members = unaccounted_for_members.value_counts(SOURCE_SPEC.PATHWAY_ID)
+    top_pathway = pathway_members[pathway_members == max(pathway_members)].index[0]
+
+    return top_pathway
+
+
+def _select_top_pathway_by_enrichment(
+    unaccounted_for_members: pd.DataFrame,
+    source_total_counts: pd.Series,
+    n_total_entities: int,
+    table_pk: str,
+    min_pw_size: int = 5,
+) -> str:
+
+    n_observed_entities = len(
+        unaccounted_for_members.index.get_level_values(table_pk).unique()
+    )
+    pathway_members = unaccounted_for_members.value_counts(
+        SOURCE_SPEC.PATHWAY_ID
+    ).rename("observed_members")
+
+    pathway_members = pathway_members.loc[pathway_members >= min_pw_size]
+    if pathway_members.shape[0] == 0:
+        return None
+
+    wide_contingency_table = (
+        pathway_members.to_frame()
+        .join(source_total_counts)
+        .assign(
+            missing_members=lambda x: x["total_counts"] - x["observed_members"],
+            observed_nonmembers=lambda x: n_observed_entities - x["observed_members"],
+            nonobserved_nonmembers=lambda x: n_total_entities
+            - x["observed_nonmembers"]
+            - x["missing_members"]
+            - x["observed_members"],
+        )
+        .drop(columns=["total_counts"])
+    )
+
+    # calculate enrichments using a fast vectorized normal approximation
+    odds_ratios, _ = hypothesis_testing.fisher_exact_vectorized(
+        wide_contingency_table["observed_members"],
+        wide_contingency_table["missing_members"],
+        wide_contingency_table["observed_nonmembers"],
+        wide_contingency_table["nonobserved_nonmembers"],
+    )
+
+    return pathway_members.index[np.argmax(odds_ratios)]
+
+
+def _update_unaccounted_for_members(
+    top_pathway, unaccounted_for_members
+) -> pd.DataFrame:
+    """
+    Update the unaccounted for members dataframe by removing the members
+    associated with the top pathway.
+
+    Parameters
+    ----------
+    top_pathway: str
+        the pathway to remove from the unaccounted for members
+    unaccounted_for_members: pd.DataFrame
+        the dataframe of unaccounted for members
+
+    Returns
+    -------
+    unaccounted_for_members: pd.DataFrame
+        the dataframe of unaccounted for members with the top pathway removed
+    """
+
+    table_type = sbml_dfs_utils.infer_entity_type(unaccounted_for_members)
+    pk = SBML_DFS_SCHEMA.SCHEMA[table_type][SCHEMA_DEFS.PK]
+
+    members_captured = (
+        unaccounted_for_members[
+            unaccounted_for_members[SOURCE_SPEC.PATHWAY_ID] == top_pathway
+        ]
+        .index.get_level_values(pk)
+        .tolist()
+    )
+
+    return unaccounted_for_members[
+        ~unaccounted_for_members.index.get_level_values(pk).isin(members_captured)
+    ]

napistu/statistics/hypothesis_testing.py

@@ -0,0 +1,66 @@
+from typing import Union
+
+import numpy as np
+from scipy.stats import norm
+
+
+def fisher_exact_vectorized(
+    observed_members: Union[list[int], np.ndarray],
+    missing_members: Union[list[int], np.ndarray],
+    observed_nonmembers: Union[list[int], np.ndarray],
+    nonobserved_nonmembers: Union[list[int], np.ndarray],
+) -> tuple[np.ndarray, np.ndarray]:
+    """
+    Fast vectorized one-tailed Fisher exact test using normal approximation.
+
+    Parameters:
+    -----------
+    observed_members, missing_members, observed_nonmembers, nonobserved_nonmembers : array-like
+        The four cells of the 2x2 contingency tables (must be non-negative)
+
+    Returns:
+    --------
+    odds_ratios : numpy array
+        Odds ratios for each test
+    p_values : numpy array
+        One-tailed p-values (tests for enrichment)
+    """
+    # Convert to numpy arrays
+    a = np.array(observed_members, dtype=float)
+    b = np.array(missing_members, dtype=float)
+    c = np.array(observed_nonmembers, dtype=float)
+    d = np.array(nonobserved_nonmembers, dtype=float)
+
+    # Check for negative values and raise error
+    if np.any((a < 0) | (b < 0) | (c < 0) | (d < 0)):
+        raise ValueError("All contingency table values must be non-negative")
+
+    # Calculate odds ratios
+    odds_ratios = np.divide(
+        a * d, b * c, out=np.full_like(a, np.inf, dtype=float), where=(b * c) != 0
+    )
+
+    # Normal approximation to hypergeometric distribution
+    n = a + b + c + d
+
+    # Avoid division by zero in expected value calculation
+    expected_a = np.divide(
+        (a + b) * (a + c), n, out=np.zeros_like(n, dtype=float), where=n != 0
+    )
+
+    # Variance calculation with protection against division by zero
+    var_a = np.divide(
+        (a + b) * (c + d) * (a + c) * (b + d),
+        n * n * (n - 1),
+        out=np.ones_like(n, dtype=float),  # Default to 1 to avoid sqrt(0)
+        where=(n > 1),
+    )
+    var_a = np.maximum(var_a, 1e-10)  # Ensure positive variance
+
+    # Continuity correction and z-score
+    z = (a - expected_a - 0.5) / np.sqrt(var_a)
+
+    # One-tailed p-value (upper tail for enrichment)
+    p_values = norm.sf(z)  # 1 - norm.cdf(z)
+
+    return odds_ratios, p_values

napistu/utils.py

@@ -14,7 +14,7 @@ import zipfile
 from contextlib import closing
 from itertools import starmap
 from textwrap import fill
-from typing import Any, List, Optional, Union
+from typing import Any, Dict, Optional, List, Union
 from urllib.parse import urlparse
 from pathlib import Path
 from requests.adapters import HTTPAdapter
@@ -1131,6 +1131,28 @@ def safe_fill(x: str, fill_width: int = 15) -> str:
         return fill(x, fill_width)
 
 
+def match_regex_dict(s: str, regex_dict: Dict[str, any]) -> Optional[any]:
+    """
+    Apply each regex in regex_dict to the string s. If a regex matches, return its value.
+    If no regex matches, return None.
+
+    Parameters
+    ----------
+    s : str
+        The string to test.
+    regex_dict : dict
+        Dictionary where keys are regex patterns (str), and values are the values to return.
+
+    Returns
+    -------
+    The value associated with the first matching regex, or None if no match.
+    """
+    for pattern, value in regex_dict.items():
+        if re.search(pattern, s):
+            return value
+    return None
+
+
 def _add_nameness_score_wrapper(df, name_var, table_schema):
     """Call _add_nameness_score with default value."""
 

{napistu-0.4.2.dist-info → napistu-0.4.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: napistu
-Version: 0.4.2
+Version: 0.4.4
 Summary: Connecting high-dimensional data to curated pathways
 Home-page: https://github.com/napistu/napistu-py
 Author: Sean Hackett

{napistu-0.4.2.dist-info → napistu-0.4.4.dist-info}/RECORD

@@ -1,13 +1,13 @@
 napistu/__init__.py,sha256=dFXAhIqlTLJMwowS4BUDT08-Vy3Q0u1L0CMCErSZT1Y,239
 napistu/__main__.py,sha256=xwlbh_0Ig3a-yG6BIJRiDPSN9R2HnX2pEBvlodlO6h4,29015
-napistu/consensus.py,sha256=xWXiqIM6ot-SSPJZXTrVpohbINSCkZXBtRi-5REfk_g,69897
+napistu/consensus.py,sha256=SDw58vkDivzy5AiOQUnf5vUbFxmSrMGMMmptDMZhk0E,69807
 napistu/constants.py,sha256=8sp1l0cxu2rsnCrWBEEwhcBKvDtc4u0D0f_72zILLW0,13427
 napistu/identifiers.py,sha256=e2-nTVzr5AINa0y1ER9218bKXyF2kAeJ9At22S4Z00o,33914
 napistu/indices.py,sha256=Zjg3gE0JQ3T879lCPazYg-WXVE6hvcAr713ZKpJ32rk,9830
 napistu/sbml_dfs_core.py,sha256=s0OyoHs-AjOcbZu1d3KNkW_PI7Rxbhu5ZLpfQeO4iY8,72639
-napistu/sbml_dfs_utils.py,sha256=w5dFcJFDKnKDK9jxPOCuCW8IccxdXmyNmP9vCUhVdf8,46184
-napistu/source.py,sha256=UGpN70bqbC9gnKmM0ivSdQYim9hfzgABeXoQKzRr9oU,13646
-napistu/utils.py,sha256=PEAsLn7VGN8JlNJQcAMYpjF1gr2mWmb5IqBsypP9hi0,35768
+napistu/sbml_dfs_utils.py,sha256=SOy1Ii2hDFOfQa7pFAJS9EfAmfBVD_sHvDJBVmCN_p8,46456
+napistu/source.py,sha256=iDDKpN-4k_W_tyxEjqe_z-yPJv7uoFRRBhkiBtOH5C8,20416
+napistu/utils.py,sha256=p2sJxTklmV30XS6hanJRjcdfgeaZpkULuMyQX3BPP0c,36404
 napistu/context/__init__.py,sha256=LQBEqipcHKK0E5UlDEg1ct-ymCs93IlUrUaH8BCevf0,242
 napistu/context/discretize.py,sha256=Qq7zg46F_I-PvQIT2_pEDQV7YEtUQCxKoRvT5Gu9QsE,15052
 napistu/context/filtering.py,sha256=l1oq-43ysSGqU9VmhTOO_pYT4DSMf20yxvktPC1MI0I,13696
@@ -17,13 +17,14 @@ napistu/gcs/downloads.py,sha256=SvGv9WYr_Vt3guzyz1QiAuBndeKPTBtWSFLj1-QbLf4,6348
 napistu/gcs/utils.py,sha256=eLSsvewWJdCguyj2k0ozUGP5BTemaE1PZg41Z3aY5kM,571
 napistu/ingestion/__init__.py,sha256=dFXAhIqlTLJMwowS4BUDT08-Vy3Q0u1L0CMCErSZT1Y,239
 napistu/ingestion/bigg.py,sha256=f65--8ARe248eYCUJpFMF284Wz53sLyFyBuwelxHmJA,4340
-napistu/ingestion/constants.py,sha256=9UP47VImZ11q0kz17N3EJg2155USqLewwNWyKpA-cbA,8089
+napistu/ingestion/constants.py,sha256=jo3v8Z7Y_tNNhTmEcokVOh1HBJFAXc-Z38S4mG58qfo,10059
 napistu/ingestion/gtex.py,sha256=X0hSC1yrpf4xSJWFhpeNcnHwJzKDII2MvjfUqYA0JN8,3720
 napistu/ingestion/hpa.py,sha256=R27ExrryKQ4Crxv9ATXmBJCa-yd01TMOrDjkeBhIQac,5054
 napistu/ingestion/identifiers_etl.py,sha256=6ppDUA6lEZurdmVbiFLOUzphYbr-hndMhtqsQnq_yAc,5009
 napistu/ingestion/napistu_edgelist.py,sha256=4RLXsoIk_-Atu-Nqme_t1JpEpBET26VIY2Y_Hcd3sMw,3580
 napistu/ingestion/obo.py,sha256=AQkIPWbjA464Lma0tx91JucWkIwLjC7Jgv5VHGRTDkE,9601
 napistu/ingestion/psi_mi.py,sha256=5eJjm7XWogL9oTyGqR52kntHClLwLsTePKqCvUGyi-w,10111
+napistu/ingestion/reactom_fi.py,sha256=hKdOY2wNtcNk6WlnHnNalryiXv6mtcWUiBW9isXPB0Y,6991
 napistu/ingestion/reactome.py,sha256=Hn9X-vDp4o_HK-OtaQvel3vJeZ8_TC1-4N2rruK9Oks,7099
 napistu/ingestion/sbml.py,sha256=l8Z98yWuOIRGns8G4UNnoQz7v_xmukZb_IZ_5ye34Ko,25296
 napistu/ingestion/string.py,sha256=go1WGTkoLJejX7GQWf9bFeInFGAw4jNSpS2B_Zr5f_s,11364
@@ -61,14 +62,14 @@ napistu/network/__init__.py,sha256=dFXAhIqlTLJMwowS4BUDT08-Vy3Q0u1L0CMCErSZT1Y,2
 napistu/network/constants.py,sha256=nG_lUZYLgop8oxOGjDYqvxXJzVdOwKZ3aWnxlhtSaIo,6915
 napistu/network/data_handling.py,sha256=KncrAKjXI3169BgVE-SnY8FkpVF60JnUwfMHtbqvsTc,14725
 napistu/network/ig_utils.py,sha256=MuyEyOVtSHndil6QuuRCimBZrJ2jTaF5qQESgYlu02M,17042
-napistu/network/neighborhoods.py,sha256=g5QeGaizSfW4nNe9YZY86g8q79EQmuvSwipaNPnOVqA,56121
+napistu/network/neighborhoods.py,sha256=kXoD5d3plcTEw-6XCbb5QjaCt0jsKwn17VdAvnGoFhY,57041
 napistu/network/net_create.py,sha256=66kV_xoWnu4BVLaJZ1TAC7wBSsjPDqjoAXH-X9ShV3s,59091
 napistu/network/net_create_utils.py,sha256=zajwaz2xAij_9fEnD77SgBw_EnNAnJ8jBCmmK2rk_bA,24672
 napistu/network/net_propagation.py,sha256=Il5nDOWh3nLz8gRhDFHGp2LxcvJ9C1twiSZjDeiZMUo,23490
 napistu/network/ng_core.py,sha256=dGnTUKR4WtnvaYMyIHqqF55FY4mJSa7wjA2LZ4cVB6U,11720
-napistu/network/ng_utils.py,sha256=c1tHXz_JcH01D5KovNQmRLTEVxpCkCe36otULq-liz8,15579
+napistu/network/ng_utils.py,sha256=ahSm-8M2pV662V7MMVcGaoguBM55_y-F7LDmZSVp9ag,15951
 napistu/network/paths.py,sha256=r6LVKVvX7i3ctBA5r-xvHfpH5Zsd0VDHUCtin2iag20,17453
-napistu/network/precompute.py,sha256=ibL0ByY7Wp5kEfIG3LUDpQKdvAeQX0DNkT_46g2YrGc,8367
+napistu/network/precompute.py,sha256=ARU2tktWnxFISaHAY8chpkg8pusZPv7TT5jSIB9eFF0,10081
 napistu/ontologies/__init__.py,sha256=dFXAhIqlTLJMwowS4BUDT08-Vy3Q0u1L0CMCErSZT1Y,239
 napistu/ontologies/constants.py,sha256=GyOFvezSxDK1VigATcruTKtNhjcYaid1ggulEf_HEtQ,4345
 napistu/ontologies/dogma.py,sha256=VVj6NKBgNym4SdOSu8g22OohALj7cbObhIJmdY2Sfy0,8860
@@ -84,8 +85,9 @@ napistu/scverse/__init__.py,sha256=Lgxr3iMQAkTzXE9BNz93CndNP5djzerLvmHM-D0PU3I,3
 napistu/scverse/constants.py,sha256=0iAkhyJUIeFGHdLLU3fCaEU1O3Oix4qAsxr3CxGTjVs,653
 napistu/scverse/loading.py,sha256=jqiE71XB-wdV50GyZrauFNY0Lai4bX9Fm2Gv80VR8t8,27016
 napistu/statistics/__init__.py,sha256=dFXAhIqlTLJMwowS4BUDT08-Vy3Q0u1L0CMCErSZT1Y,239
+napistu/statistics/hypothesis_testing.py,sha256=k0mBFAMF0XHVcKwS26aPnEbq_FIUVwXU1gZ6cKfFbCk,2190
 napistu/statistics/quantiles.py,sha256=1-LnmVzC2CQWxCKUh0yi6YfKrbsZM1-kkD7nu2-aS5s,3042
-napistu-0.4.2.dist-info/licenses/LICENSE,sha256=kW8wVT__JWoHjl2BbbJDAZInWa9AxzJeR_uv6-i5x1g,1063
+napistu-0.4.4.dist-info/licenses/LICENSE,sha256=kW8wVT__JWoHjl2BbbJDAZInWa9AxzJeR_uv6-i5x1g,1063
 tests/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
 tests/conftest.py,sha256=t-GHb0MvSsC-MyhkFpOy2K3t5fi7eaig_Rc2xEQC-t8,9678
 tests/test_consensus.py,sha256=Hzfrgp4SpkRDnEMVMD3f0UInSycndB8kKzC4wDDvRas,15076
@@ -114,7 +116,7 @@ tests/test_network_net_propagation.py,sha256=kZeDHD93iMrLVvxO4OyfRH5_vgsYeQyC40O
 tests/test_network_ng_core.py,sha256=w-iNBTtenennJhaLFauk952pEsk7W0-Fa8lPvIRqHyY,628
 tests/test_network_ng_utils.py,sha256=QVVuRnvCRfTSIlGdwQTIF9lr0wOwoc5gGeXAUY_AdgE,713
 tests/test_network_paths.py,sha256=TWZnxY5bF3m6gahcxcYJGrBIawh2-_vUcec1LyPmXV8,1686
-tests/test_network_precompute.py,sha256=zwJrKNC3s8rIrsyAQfQMYxbl8HZXUr7u09nMJ_K8jiU,9005
+tests/test_network_precompute.py,sha256=IPr1KhtxBD0fXx_2TvZqnevrD-Iig35otb8yloRFpRc,10014
 tests/test_ontologies_genodexito.py,sha256=6fINyUiubHZqu7qxye09DQfJXw28ZMAJc3clPb-cCoY,2298
 tests/test_ontologies_id_tables.py,sha256=CpwpbmQvTc1BaVd6jbDKHAVE2etwN0vx93nC8jpnMlE,7265
 tests/test_ontologies_mygene.py,sha256=VkdRcKIWmcG6V-2dpfvsBiOJN5dO-j0RqZNxtJRcyBU,1583
@@ -124,18 +126,18 @@ tests/test_rpy2_callr.py,sha256=V4a-QH5krgYOQRgqzksMzIkGAFjBqKOAqgprxrH6bE0,2904
 tests/test_rpy2_init.py,sha256=T3gnxC1O7XNvYM2P4018ikpPPAy-kwQLm7Erj0RfA-4,5895
 tests/test_sbml.py,sha256=f25zj1NogYrmLluvBDboLameTuCiQ309433Qn3iPvhg,1483
 tests/test_sbml_dfs_core.py,sha256=nnLPpZTVtCznOBohk7CX67x6sMqktJWt-sZMWQKoaDs,26521
-tests/test_sbml_dfs_utils.py,sha256=gWIhzUEtQlOR9c1TiCyhlSAELmWnBSncn6vCEqH5hl0,11029
+tests/test_sbml_dfs_utils.py,sha256=ZD9x2B81fsfYEjAV9wphHOR7ywjNcfvfw1LGNv4PxUA,11471
 tests/test_sbo.py,sha256=x_PENFaXYsrZIzOZu9cj_Wrej7i7SNGxgBYYvcigLs0,308
 tests/test_scverse_loading.py,sha256=bnU1lQSYYWhOAs0IIBoi4ZohqPokDQJ0n_rtkAfEyMU,29948
-tests/test_set_coverage.py,sha256=J-6m6LuOjcQa9pxRuWglSfJk4Ltm7kt_eOrn_Q-7P6Q,1604
-tests/test_source.py,sha256=hT0IlpexR5zP0OhWl5BBaho9d1aCYQlFZLwRIRRnw_Y,1969
+tests/test_source.py,sha256=iV-Yyu8flhIGWF17SCL8msG2bjqwb9w2IZ694b0iZ-o,2985
+tests/test_statistics_hypothesis_testing.py,sha256=qD-oS9zo5JlH-jdtiOrWAKI4nKFuZvvh6361_pFSpIs,2259
 tests/test_statistics_quantiles.py,sha256=yNDeqwgbP-1Rx3C_dLX_wnwT_Lr-iJWClmeKmElqmTE,4984
 tests/test_uncompartmentalize.py,sha256=nAk5kfAVLU9a2VWe2x2HYVcKqj-EnwmwddERIPRax8c,1289
 tests/test_utils.py,sha256=qPSpV-Q9b6vmdycgaDmQqtcvzKnAVnN9j5xJ9x-T6bg,23959
 tests/utils.py,sha256=SoWQ_5roJteFGcMaOeEiQ5ucwq3Z2Fa3AAs9iXHTsJY,749
 tests/test_data/__init__.py,sha256=47DEQpj8HBSa-_TImW-5JCeuQeRkm5NMpJWZG3hSuFU,0
-napistu-0.4.2.dist-info/METADATA,sha256=6P_9Mmno6pVu4Me-3QdcMtiGOhCcajTqm5LP_Hns4lI,4078
-napistu-0.4.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-napistu-0.4.2.dist-info/entry_points.txt,sha256=_QnaPOvJNA3IltxmZgWIiBoen-L1bPYX18YQfC7oJgQ,41
-napistu-0.4.2.dist-info/top_level.txt,sha256=Gpvk0a_PjrtqhYcQ9IDr3zR5LqpZ-uIHidQMIpjlvhY,14
-napistu-0.4.2.dist-info/RECORD,,
+napistu-0.4.4.dist-info/METADATA,sha256=E15A5Ve2RZTn4HtXGD2rDO1Q7AEaTfSdo3fgLuwravE,4078
+napistu-0.4.4.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+napistu-0.4.4.dist-info/entry_points.txt,sha256=_QnaPOvJNA3IltxmZgWIiBoen-L1bPYX18YQfC7oJgQ,41
+napistu-0.4.4.dist-info/top_level.txt,sha256=Gpvk0a_PjrtqhYcQ9IDr3zR5LqpZ-uIHidQMIpjlvhY,14
+napistu-0.4.4.dist-info/RECORD,,

tests/test_network_precompute.py

@@ -276,3 +276,33 @@ def test_precomputed_distances_serialization():
         # Clean up the temporary file
         if os.path.exists(temp_path):
             os.remove(temp_path)
+
+
+def test_filter_precomputed_distances_top_n_subset():
+    # Use a small top_n for a quick test
+    top_n = 5
+    filtered = precompute.filter_precomputed_distances_top_n(
+        precomputed_distances, top_n=top_n
+    )
+    # Check that the filtered DataFrame is a subset of the original
+    merged = filtered.merge(
+        precomputed_distances,
+        on=[
+            precompute.NAPISTU_EDGELIST.SC_ID_ORIGIN,
+            precompute.NAPISTU_EDGELIST.SC_ID_DEST,
+        ],
+        how="left",
+        indicator=True,
+    )
+    assert (
+        merged["_merge"] == "both"
+    ).all(), "Filtered rows must be present in the original DataFrame"
+    # Check that columns are preserved
+    assert set(
+        [
+            precompute.NAPISTU_EDGELIST.SC_ID_ORIGIN,
+            precompute.NAPISTU_EDGELIST.SC_ID_DEST,
+        ]
+    ).issubset(filtered.columns)
+    # Optionally, check that the number of rows is less than or equal to the input
+    assert filtered.shape[0] <= precomputed_distances.shape[0]

tests/test_sbml_dfs_utils.py

@@ -334,3 +334,16 @@ def test_infer_entity_type_errors():
     )  # Two primary keys
     with pytest.raises(ValueError):
         sbml_dfs_utils.infer_entity_type(df)
+
+
+def test_infer_entity_type_multindex_reactions():
+    # DataFrame with MultiIndex (r_id, foo), should infer as reactions
+    import pandas as pd
+    from napistu.constants import SBML_DFS
+
+    df = pd.DataFrame({"some_col": [1, 2]})
+    df.index = pd.MultiIndex.from_tuples(
+        [("rxn1", "a"), ("rxn2", "b")], names=[SBML_DFS.R_ID, "foo"]
+    )
+    result = sbml_dfs_utils.infer_entity_type(df)
+    assert result == SBML_DFS.REACTIONS