reorg src dir; bring back gene clustering; add new cell clustering functions

Author: pcahan1

docs/CHANGELOG.md

@@ -4,9 +4,28 @@ All notable changes to PySingleCellNet should be listed here. The definition of
 
 ## [Unreleased]
 
+### Changed
+
+- replace cl with tl
+- moved functions to more fitting files, like unused ones to utils.misc.py
+- do not export unused functions
+
+### Added
+
+- tl.discover_cell_cliques labels cells by consensus cluster labels, kind of
+- tl.clustering_quality_vs_nn_summary computes metrics of clustering quality
+- tl.cluster_alot
+- resurrected gene clustering functions
+
+### Fixed
+
+- `filter_anndata_slots` to handle .uns and dependencies across slots
+
 ### Removed
 
-- lots o stuff that is old or has been moved to STUF
+- ut.mito_rib_heme
+- 
+- lots of stuff that is old or has been moved to other packages like STUF
 
 ## [0.1.2] - 2025-08-05
 

src/pySingleCellNet/__init__.py

@@ -4,14 +4,14 @@
 from .config import SCN_DIFFEXP_KEY
 from . import plotting as pl
 from . import utils as ut
-from . import classify as cl
+from . import tools as tl
 
 # Public API
 __all__ = [
     "__version__",
     "pl",
     "ut",
-    "cl"
+    "tl"
 ]    
 
 

src/pySingleCellNet/classify/__init__.py src/pySingleCellNet/tools/__init__.pysrc/pySingleCellNet/classify/__init__.py renamed to src/pySingleCellNet/tools/__init__.py

@@ -1,7 +1,19 @@
+from .cluster import (
+    cluster_alot,
+    cluster_subclusters,
+)
+
+from .cluster_eval import (
+    clustering_quality_vs_nn_summary
+)
+
+from .cluster_cliques import ( 
+    discover_cell_cliques
+)
+
 from .classifier import (
     classify_anndata,
     train_classifier,
-    train_and_assess,
     create_classifier_report
 )
 
@@ -19,11 +31,23 @@
     deg
 )
 
+from .gene import (
+    build_gene_knn,
+    find_gene_modules,
+    whoare_genes_neighbors,
+    score_gene_modules,
+    what_module_has_gene,
+)
+
+
 # API
 __all__ = [
+    "cluster_alot",
+    "cluster_subcluster",
+    "clustering_quality_vs_nn_summary",
+    "discover_cell_cliques",
     "classify_anndata",
     "train_classifier",
-    "train_and_assess",
     "create_classifier_report",
     "categorize_classification",
     "comp_ct_thresh",
@@ -32,6 +56,11 @@
     "gsea_on_deg",
     "collect_gsea_results_from_dict",
     "convert_diffExp_to_dict",
-    "deg"
+    "deg",
+    "build_gene_knn",
+    "find_gene_modules",
+    "whoare_genes_neighbors",
+    "score_gene_modules",
+    "what_module_has_gene",
 ]
 

…c/pySingleCellNet/classify/categorize.py src/pySingleCellNet/tools/categorize.pysrc/pySingleCellNet/classify/categorize.py renamed to src/pySingleCellNet/tools/categorize.py src/pySingleCellNet/classify/categorize.py renamed to src/pySingleCellNet/tools/categorize.py

@@ -3,8 +3,7 @@
 import scanpy as sc
 import anndata
 from anndata import AnnData
-from typing import List
-from typing import Dict
+from typing import List, Dict
 from sklearn.ensemble import RandomForestClassifier
 from scipy.sparse import csr_matrix
 import warnings
@@ -13,11 +12,52 @@
 #from ..utils import *
 #from .tsp_rf import *
 #from .scn_assess import create_classifier_report
-from ..utils import build_knn_graph, rank_genes_subsets, get_unique_colors, split_adata_indices
+from ..utils import build_knn_graph, get_unique_colors, split_adata_indices
 from sklearn.metrics import classification_report
 from pySingleCellNet.config import SCN_DIFFEXP_KEY, SCN_CATEGORY_COLOR_DICT
 import random as rand 
 
+def _rank_genes_subsets(
+    adata,
+    groupby,
+    grpA,
+    grpB,
+    pval = 0.01,
+    layer=None
+):
+    """
+    Subset an AnnData object to specified groups, create a new .obs column labeling cells
+    as group A or B, and run rank_genes_groups for differential expression analysis. Necessary because the scanpy reference does not seem to work
+    
+    Parameters:
+        adata (AnnData): The AnnData object.
+        groupby (str): The .obs column to group cells by.
+        grpA (list): Values used to subset cells into group A.
+        grpB (list): Values used to subset cells into group B.
+        layer (str, optional): Layer to use for expression values.
+        
+    Returns:
+        AnnData: Subsetted and labeled AnnData object after running rank_genes_groups.
+    """
+    # Subset the data to cells in either grpA or grpB
+    subset = adata[adata.obs[groupby].isin(grpA + grpB)].copy()
+    # Create a new .obs column labeling cells as 'grpA' or 'grpB'
+    subset.obs["comparison_group"] = subset.obs[groupby].apply(
+        lambda x: "grpA" if x in grpA else "grpB"
+    )
+    # Run rank_genes_groups
+    sc.tl.rank_genes_groups(
+        subset,
+        groupby="comparison_group",
+        layer=layer,
+        pts = True,
+        use_raw=False
+    )
+    # return subset
+    ans = sc.get.rank_genes_groups_df(subset, group='grpA', pval_cutoff=pval)
+    return ans
+
+
 def _query_transform(expMat, genePairs):
     npairs = len(genePairs)
     ans = pd.DataFrame(0, index = expMat.index, columns = np.arange(npairs))
@@ -254,7 +294,7 @@ def _get_classy_genes_3(
         ]["name"]
 
         xdata = adata.copy()
-        subsetDF = rank_genes_subsets(
+        subsetDF = _rank_genes_subsets(
             xdata, groupby=groupby, grpA=[g], grpB=other_groups, layer=layer, pval = pval
         )
 
@@ -294,7 +334,7 @@ def _get_classy_genes_3(
             all_others = [x for x in groups if x != g]
 
             # E.g., run a fallback differential expression
-            fallbackDF = rank_genes_subsets(
+            fallbackDF = _rank_genes_subsets(
                 xdata_fallback, groupby=groupby, grpA=[g], grpB=all_others, layer=layer, pval=1
             )
             fallback_genes = get_top_genes_from_df(
@@ -684,7 +724,7 @@ def pick_different_gene(current_gene, gene_pool):
     # Convert to a NumPy array of unique values
     return np.unique(all_pairs)
 
-
+# deprecated 
 def train_and_assess(
     adata,
     groupby,
@@ -718,3 +758,5 @@ def train_and_assess(
 
 
 
+
+