risk-network 0.0.9b28__py3-none-any.whl → 0.0.9b29__py3-none-any.whl

risk/__init__.py

@@ -7,4 +7,4 @@ RISK: Regional Inference of Significant Kinships
 
 from risk.risk import RISK
 
-__version__ = "0.0.9-beta.28"
+__version__ = "0.0.9-beta.29"

risk/stats/stat_tests.py

@@ -3,6 +3,7 @@ risk/stats/stat_tests
 ~~~~~~~~~~~~~~~~~~~~~
 """
 
+from concurrent.futures import ThreadPoolExecutor
 from typing import Any, Dict
 
 import numpy as np
@@ -120,56 +121,62 @@ def compute_hypergeom_test(
     annotations: csr_matrix,
     null_distribution: str = "network",
 ) -> Dict[str, Any]:
-    """
-    Compute hypergeometric test for enrichment and depletion in neighborhoods with selectable null distribution.
+    """Compute hypergeometric test for enrichment and depletion in neighborhoods with selectable null distribution.
 
     Args:
         neighborhoods (csr_matrix): Sparse binary matrix representing neighborhoods.
         annotations (csr_matrix): Sparse binary matrix representing annotations.
-        null_distribution (str, optional): Type of null distribution ('network' or 'annotations'). Defaults to "network".
+        null_distribution (str, optional): Type of null distribution ('network' or 'annotations').
+            Defaults to "network".
 
     Returns:
         Dict[str, Any]: Dictionary containing depletion and enrichment p-values.
     """
-    # Total number of nodes
+    # Get the total number of nodes in the network
     total_nodes = neighborhoods.shape[1]
+    # Calculate neighborhood and annotation sums
+    neighborhood_sums = neighborhoods.sum(axis=0).A1  # A1 returns a 1D array
+    annotation_sums = annotations.sum(axis=0).A1
 
-    # Compute sums directly using sparse operations
-    neighborhood_sums = neighborhoods.sum(axis=0).A.flatten()
-    annotation_sums = annotations.sum(axis=0).A.flatten()
-
-    if null_distribution == "network":
-        background_population = total_nodes
-    elif null_distribution == "annotations":
-        # Boolean mask for nodes with annotations
-        annotated_nodes = annotations.getnnz(axis=1) > 0
+    if null_distribution == "annotations":
+        annotated_nodes = annotations.getnnz(axis=1) > 0  # Nodes with any annotation
         background_population = annotated_nodes.sum()
-        # Filter neighborhoods and annotations to include only annotated nodes
         neighborhoods = neighborhoods[annotated_nodes]
         annotations = annotations[annotated_nodes]
-        neighborhood_sums = neighborhoods.sum(axis=0).A.flatten()
-        annotation_sums = annotations.sum(axis=0).A.flatten()
+        neighborhood_sums = neighborhoods.sum(axis=0).A1
+        annotation_sums = annotations.sum(axis=0).A1
+    elif null_distribution == "network":
+        background_population = total_nodes
     else:
-        raise ValueError(
-            "Invalid null_distribution value. Choose either 'network' or 'annotations'."
+        raise ValueError("Invalid null_distribution value. Choose 'network' or 'annotations'.")
+
+    # Sparse matrix multiplication for observed counts
+    annotated_in_neighborhood = neighborhoods.T @ annotations  # Result is sparse
+
+    def compute_pvals_for_column(col_idx: int):
+        """Compute depletion and enrichment p-values for a single annotation column."""
+        observed_counts = annotated_in_neighborhood[:, col_idx].toarray().flatten()
+        ann_total = annotation_sums[col_idx]
+        # Compute depletion and enrichment p-values
+        depletion_pvals = hypergeom.cdf(
+            observed_counts, background_population, ann_total, neighborhood_sums
         )
-
-    # Compute annotated nodes in each neighborhood
-    annotated_in_neighborhood = neighborhoods.T @ annotations  # Sparse multiplication
-    # Convert to dense arrays for vectorized operations
-    annotated_in_neighborhood = annotated_in_neighborhood.toarray()
-    # Align shapes for broadcasting
-    neighborhood_sums = neighborhood_sums[:, np.newaxis]
-    annotation_sums = annotation_sums[np.newaxis, :]
-    background_population = np.array([[background_population]])
-
-    # Fully vectorized hypergeometric calculations
-    depletion_pvals = hypergeom.cdf(
-        annotated_in_neighborhood, background_population, annotation_sums, neighborhood_sums
-    )
-    enrichment_pvals = hypergeom.sf(
-        annotated_in_neighborhood - 1, background_population, annotation_sums, neighborhood_sums
-    )
+        enrichment_pvals = hypergeom.sf(
+            observed_counts - 1, background_population, ann_total, neighborhood_sums
+        )
+        return col_idx, depletion_pvals, enrichment_pvals
+
+    # Use ThreadPoolExecutor to process columns in parallel
+    num_columns = annotations.shape[1]
+    depletion_pvals = np.empty((annotated_in_neighborhood.shape[0], num_columns), dtype=np.float64)
+    enrichment_pvals = np.empty((annotated_in_neighborhood.shape[0], num_columns), dtype=np.float64)
+    with ThreadPoolExecutor() as executor:
+        results = executor.map(compute_pvals_for_column, range(num_columns))
+
+    # Collect results
+    for col_idx, dep_pval, enr_pval in results:
+        depletion_pvals[:, col_idx] = dep_pval
+        enrichment_pvals[:, col_idx] = enr_pval
 
     return {"depletion_pvals": depletion_pvals, "enrichment_pvals": enrichment_pvals}
 

{risk_network-0.0.9b28.dist-info → risk_network-0.0.9b29.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.2
 Name: risk-network
-Version: 0.0.9b28
+Version: 0.0.9b29
 Summary: A Python package for biological network analysis
 Author: Ira Horecka
 Author-email: Ira Horecka <ira89@icloud.com>

{risk_network-0.0.9b28.dist-info → risk_network-0.0.9b29.dist-info}/RECORD

@@ -1,4 +1,4 @@
-risk/__init__.py,sha256=32Lq_wPcVY8stW7c0jkvgihM15jnYka5Hnw8M9gbjN0,127
+risk/__init__.py,sha256=zxCDSH8dgk6s0jbXYvLcOxAb7IGjNoZoOCoCwQ1Vd7g,127
 risk/constants.py,sha256=XInRaH78Slnw_sWgAsBFbUHkyA0h0jL0DKGuQNbOvjM,550
 risk/risk.py,sha256=s827_lRknFseOP9O4zW8sP-IcCd2EzrpV_tnVY_tz5s,1104
 risk/annotations/__init__.py,sha256=parsbcux1U4urpUqh9AdzbDWuLj9HlMidycMPkpSQFo,179
@@ -30,12 +30,12 @@ risk/network/plotter/utils/colors.py,sha256=VU1sLPRC99ll6EGK4vRNgLMUXU8lja1vjiXU
 risk/network/plotter/utils/layout.py,sha256=OPqV8jzV9dpnOhYU4SYMSfsIXalVzESrlBSI_Y43OGU,3640
 risk/stats/__init__.py,sha256=2zdLv3tUHKyAjwAo7LprVXRaak1cHgrpYMVMSik6JM4,324
 risk/stats/significance.py,sha256=6cKv2xBQXWTHZ6HpNWIqlNfKKS5pG_BcCUdMM3r_zw4,7336
-risk/stats/stat_tests.py,sha256=qYn85VrNJeIlEptkEUoYsPco4BQ604CLJxXczgekXgc,11986
+risk/stats/stat_tests.py,sha256=kZu48JPDneMwAZGVci2i_oaJUPLUKwiYTT3c4pKhr-0,12490
 risk/stats/permutation/__init__.py,sha256=OLmYLm2uj96hPsSaUs0vUqFYw6Thwch_aHtpL7L0ZFw,127
 risk/stats/permutation/permutation.py,sha256=BWjgdBpLVcHvmwHy0bmD4aJFccxifNBSrrCBPppyKf4,10569
 risk/stats/permutation/test_functions.py,sha256=D3XMPM8CasUNytWSRce22TI6KK6XulYn5uGG4lWxaHs,3120
-risk_network-0.0.9b28.dist-info/LICENSE,sha256=jOtLnuWt7d5Hsx6XXB2QxzrSe2sWWh3NgMfFRetluQM,35147
-risk_network-0.0.9b28.dist-info/METADATA,sha256=v9uuUQ9EwyI5WzIirw_ONry2KPaiHqTiw2TDWX60Y6c,47627
-risk_network-0.0.9b28.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
-risk_network-0.0.9b28.dist-info/top_level.txt,sha256=NX7C2PFKTvC1JhVKv14DFlFAIFnKc6Lpsu1ZfxvQwVw,5
-risk_network-0.0.9b28.dist-info/RECORD,,
+risk_network-0.0.9b29.dist-info/LICENSE,sha256=jOtLnuWt7d5Hsx6XXB2QxzrSe2sWWh3NgMfFRetluQM,35147
+risk_network-0.0.9b29.dist-info/METADATA,sha256=K9e3VRGdXrMUzmhtpX0PLQBFCdakxyBgKbW88xpIVYw,47627
+risk_network-0.0.9b29.dist-info/WHEEL,sha256=In9FTNxeP60KnTkGw7wk6mJPYd_dQSjEZmXdBdMCI-8,91
+risk_network-0.0.9b29.dist-info/top_level.txt,sha256=NX7C2PFKTvC1JhVKv14DFlFAIFnKc6Lpsu1ZfxvQwVw,5
+risk_network-0.0.9b29.dist-info/RECORD,,