cluster_alot etc

Author: pcahan1

docs/CHANGELOG.md

@@ -4,6 +4,8 @@ All notable changes to PySingleCellNet should be listed here. The definition of
 
 ## [Unreleased]
 
+## [0.1.3] - 2025-09-15
+
 ### Changed
 
 - replace cl with tl
@@ -15,16 +17,18 @@ All notable changes to PySingleCellNet should be listed here. The definition of
 - 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
+- tl.cluster_subcluster
+- resurrected gene clustering functions 
+- notebook tutorial on cell clustering
 
 ### Fixed
 
 - `filter_anndata_slots` to handle .uns and dependencies across slots
+- `classify_anndata` bug that prevented writing h5ad. see https://github.com/CahanLab/PySingleCellNet/issues/13
 
 ### Removed
 
 - 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

docs/classifier.md

@@ -1,10 +1,10 @@
-The easiest way to install PySingleCellNet is available on PyPI. Install it from there with:
+PySingleCellNet is available on PyPI. You can install it with:
 
 ```
 pip install pySingleCellNet
 ```
 
-Alternatively, you can install directly from GitHub with: 
+You could also install it directly from GitHub with: 
 
 ```shell
 pip install git+https://github.com/CahanLab/pySingleCellNet.git

docs/install.md

@@ -0,0 +1,7 @@
+# analysis tools including classification
+
+Functions that create or require the classifier `clf` object, and friends
+
+::: pySingleCellNet.tools
+
+

docs/notebooks/categorize.ipynb

@@ -63,6 +63,8 @@ nav:
   - Tutorials: 
     - 'Quickstart': notebooks/quickstart.ipynb
     - 'Explore': notebooks/categorize.ipynb
+    - 'Clustering cells': notebooks/cluster_alot.ipynb
+#    - 'Clustering genes': notebooks/gene_modules.ipynb
 #    - 'train & classify': 'notebooks/Basics.ipynb'
 #    - 'analyze _in vitro_ cells': 'notebooks/explore.ipynb'
 #    - 'prepare training data': 'notebooks/how-to_prepare_reference_data.ipynb'
@@ -71,9 +73,9 @@ nav:
   - Training data: training_data.md
   - References: refs.md
   - API:
-    - 'Utilities': 'utils.md'
-    - 'Classifier': 'classifier.md'
-    - 'Plotting': 'plotting.md'
+    - 'Utilities': utils.md
+    - 'Tools': tools.md
+    - 'Plotting': plotting.md
   - Release Notes: CHANGELOG.md
 
 

docs/notebooks/cluster_alot.ipynb

@@ -21,13 +21,12 @@
 from .dot import (    
     umi_counts_ranked,
     ontogeny_graph,
-    dotplot_deg,
-    dotplot_diff_gene,
     dotplot_scn_scores,
     umap_scores,
 )
 
 from .heatmap import (
+    heatmap_clustering_eval,
     heatmap_classifier_report,
     heatmap_scores,
     heatmap_gsea,
@@ -53,10 +52,9 @@
     "bar_classifier_f1",
     "umi_counts_ranked",
     "ontogeny_graph",
-    "dotplot_deg",
-    "dotplot_diff_gene",
     "dotplot_scn_scores",
     "umap_scores",
+    "heatmap_clustering_eval",
     "heatmap_classifier_report",
     "heatmap_scores",
     "heatmap_gsea",

docs/notebooks/quickstart.ipynb

@@ -8,23 +8,13 @@
 import umap
 import anndata as ad
 import igraph as ig
-# from palettable.colorbrewer.qualitative import Set2_6
-# from palettable.tableau import GreenOrange_6
-# from palettable.cartocolors.qualitative import Safe_6
-# from palettable.cartocolors.qualitative import Vivid_4
-# from palettable.cartocolors.qualitative import Vivid_6
-# from palettable.cartocolors.qualitative import Vivid_10
-# from palettable.scientific.diverging import Roma_20
-from palettable.scientific.sequential import LaJolla_20
-from palettable.scientific.sequential import Batlow_20
+from palettable.scientific.sequential import LaJolla_20, Batlow_20
 from anndata import AnnData
 from scipy.sparse import csr_matrix
 from scipy import sparse
 from typing import Optional, Callable
 from ..utils import *
 
-
-
 def umi_counts_ranked(adata, total_counts_column="total_counts"):
     """
     Identifies and plors the knee point of the UMI count distribution in an AnnData object.
@@ -117,55 +107,7 @@ def ontogeny_graph(gra, color_dict):
     ig.plot(gra, **v_style, vertex_size=node_sizes)
     plt.show()
 
-def dotplot_deg(
-    adata: AnnData,
-    diff_gene_dict: dict,
-    #samples_obsvals: list = [],
-    groupby_obsname: str = "comb_sampname", # I think that this is the column that splits cells for DEG
-    cellgrp_obsname: str = "comb_cellgrp",    # I think that this is the column that splits into discint clusters or cell types
-    cellgrp_obsvals: list = [], # this is the subset of clusters or cell types to limit this visualization to
-    num_genes: int = 10, 
-    order_by = 'scores',
-    new_obsname = 'grp_by_samp',
-    use_raw=False
-):
 
-    # remove celltypes unspecified in diff_gene_dict
-    dd_dict = diff_gene_dict['geneTab_dict']
-    tokeep = list(dd_dict.keys())
-
-    # also remove cell_types not listed in celltype_names
-    # default for celltype_names is all celltypes included in dd_dict
-    if len(cellgrp_obsvals) > 0:
-        cellgrp_obsvals = list(set(cellgrp_obsvals).intersection(set(tokeep)))
-    else:
-        cellgrp_obsvals = tokeep
-
-    adNew = adata.copy()
-    adNew = adNew[adNew.obs[cellgrp_obsname].isin(cellgrp_obsvals)].copy()
-    
-    # remove categories unspecified in diff_gene_dict
-    dictkey = list(diff_gene_dict.keys())[0]
-    sample_names = diff_gene_dict[dictkey]
-    adNew = adNew[adNew.obs[groupby_obsname].isin(sample_names)].copy()
-
-    # add column 'grp_by_samp' to obs that indicates cellgrp by sample
-    adNew.obs[new_obsname] = adNew.obs[cellgrp_obsname].astype(str) + "_X_" + adNew.obs[groupby_obsname].astype(str)
-    
-    # define dict of marker genes based on threshold
-    genes_to_plot = dict()
-    for cellgrp in cellgrp_obsvals:
-        print(f"{cellgrp}")
-        for sname in sample_names:
-            print(f"{sname}")
-            genes_to_plot[cellgrp + "_X_" + sname] = pull_out_genes_v2(diff_gene_dict, cell_type = cellgrp, category = sname, num_genes = num_genes, order_by=order_by) 
-
-    # return adNew, genes_to_plot
-    plt.rcParams['figure.constrained_layout.use'] = True
-    #xplot = sc.pl.DotPlot(adNew, genes_to_plot, 'ct_by_cat', cmap='RdPu', var_group_rotation = 0) #, dendrogram=True,ax=ax2, show=False)
-    xplot = sc.pl.DotPlot(adNew, genes_to_plot, new_obsname, cmap=LaJolla_20.mpl_colormap, var_group_rotation = 0, use_raw=False) #, dendrogram=True,ax=ax2, show=False)
-    xplot.swap_axes(True) # see sc.pl.DotPlot docs for useful info
-    return xplot
 
 
 def dotplot_diff_gene(
@@ -229,15 +171,6 @@ def dotplot_scn_scores(
     adTemp.obs[groupby] = adata.obs[groupby]
     sc.pl.dotplot(adTemp, adTemp.var_names.values, groupby=groupby, expression_cutoff=expression_cutoff, cmap=Batlow_20.mpl_colormap, colorbar_title="SCN score")
 
-def umap_scores_old(
-    adata: AnnData,
-    scn_classes: list,
-    obsm_name = 'SCN_score'
-):    
-    adTemp = AnnData(adata.obsm[obsm_name], obs=adata.obs)
-    adTemp.obsm['X_umap'] = adata.obsm['X_umap'].copy()
-    sc.pl.umap(adTemp,color=scn_classes, alpha=.75, s=10, vmin=0, vmax=1)
-
 
 def umap_scores(
     adata: AnnData,