cellSP 2.2__tar.gz → 2.2.3__tar.gz

{cellsp-2.2 → cellsp-2.2.3}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cellSP
-Version: 2.2
+Version: 2.2.3
 Summary: cellSP.
 Author-email: Bhavay Aggarwal <baggarwal9@gatech.edu>
 License-Expression: MIT
@@ -38,6 +38,10 @@ Requires-Dist: umap-learn>=0.5.7
 Dynamic: license-file
 
 # CellSP
+[![PyPI version](https://img.shields.io/pypi/v/cellSP.svg)](https://pypi.org/project/cellSP/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![Docs](https://img.shields.io/badge/docs-available-brightgreen)](https://github.com/bhavaygg/cellSP#readme)
+[![Downloads](https://static.pepy.tech/badge/cellSP)](https://pepy.tech/project/cellSP)
 
 CellSP is a python package for the analysis of subcellular spatial transcriptomic data. CellSP works with datasets generated at single-modulecule resolution from technologies like Xenium, CosMx, MERSCOPE or other ISH-like data. Using existing tools [InSTAnT](https://github.com/bhavaygg/InSTAnT) and [SPRAWL](https://github.com/salzman-lab/SPRAWL/), CellSP identifies statistically signficant subcellular patterns of gene transcripts and uses a biclustering algorithm to aggregate these patterns over hundereds of cells to produce "gene-cell modules". These modules represent the consistent detection of the same subcellular pattern by a set of genes in the same cells and offer a summarized and biologically interpretable desciption of subcellular patterns. CellSP provides specialized techniques for visualizing such modules and their defining spatial patterns. Additionally, CellSP utilize Gene Ontology (GO) enrichments tests to offer functionsal insights into the genes comprising the module as CellSPll as the cells comprising the module.
 
@@ -165,11 +169,13 @@ InSTAnT identifies colocalized gene pairs based on their spatial proximity. A pr
 
 ```
 @article{aggarwal2025cellsp,
-  title={CellSP: Module discovery and visualization for subcellular spatial transcriptomics data},
+  title={CellSP enables module discovery and visualization for subcellular spatial transcriptomics data},
   author={Aggarwal, Bhavay and Sinha, Saurabh},
-  journal={bioRxiv},
-  pages={2025--01},
+  journal={Communications Biology},
+  volume={8},
+  number={1},
+  pages={1528},
   year={2025},
-  publisher={Cold Spring Harbor Laboratory}
+  publisher={Nature Publishing Group UK London}
 }
 ```

{cellsp-2.2 → cellsp-2.2.3}/README.md

@@ -1,4 +1,8 @@
 # CellSP
+[![PyPI version](https://img.shields.io/pypi/v/cellSP.svg)](https://pypi.org/project/cellSP/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![Docs](https://img.shields.io/badge/docs-available-brightgreen)](https://github.com/bhavaygg/cellSP#readme)
+[![Downloads](https://static.pepy.tech/badge/cellSP)](https://pepy.tech/project/cellSP)
 
 CellSP is a python package for the analysis of subcellular spatial transcriptomic data. CellSP works with datasets generated at single-modulecule resolution from technologies like Xenium, CosMx, MERSCOPE or other ISH-like data. Using existing tools [InSTAnT](https://github.com/bhavaygg/InSTAnT) and [SPRAWL](https://github.com/salzman-lab/SPRAWL/), CellSP identifies statistically signficant subcellular patterns of gene transcripts and uses a biclustering algorithm to aggregate these patterns over hundereds of cells to produce "gene-cell modules". These modules represent the consistent detection of the same subcellular pattern by a set of genes in the same cells and offer a summarized and biologically interpretable desciption of subcellular patterns. CellSP provides specialized techniques for visualizing such modules and their defining spatial patterns. Additionally, CellSP utilize Gene Ontology (GO) enrichments tests to offer functionsal insights into the genes comprising the module as CellSPll as the cells comprising the module.
 
@@ -126,11 +130,13 @@ InSTAnT identifies colocalized gene pairs based on their spatial proximity. A pr
 
 ```
 @article{aggarwal2025cellsp,
-  title={CellSP: Module discovery and visualization for subcellular spatial transcriptomics data},
+  title={CellSP enables module discovery and visualization for subcellular spatial transcriptomics data},
   author={Aggarwal, Bhavay and Sinha, Saurabh},
-  journal={bioRxiv},
-  pages={2025--01},
+  journal={Communications Biology},
+  volume={8},
+  number={1},
+  pages={1528},
   year={2025},
-  publisher={Cold Spring Harbor Laboratory}
+  publisher={Nature Publishing Group UK London}
 }
 ```

{cellsp-2.2 → cellsp-2.2.3}/cellSP/characterize/_instant.py

@@ -49,6 +49,8 @@ def run_instant(adata_st, distance_threshold, threads, n_vertices = None, alpha_
             obj.run_ProximalPairs3D_slice(distance_threshold = distance_threshold, min_genecount = 20)
         else:
             df_temp = adata_st.uns["transcripts"].copy()
+            if 'absZ_raw' not in adata_st.uns["transcripts"].columns:
+                adata_st.uns["transcripts"]['absZ_raw'] = adata_st.uns["transcripts"]['absZ'].copy()
             adata_st.uns["transcripts"]['absZ'] = adata_st.uns["transcripts"]['absZ_raw']
             obj.run_ProximalPairs3D(distance_threshold = distance_threshold, min_genecount = 20)
             adata_st.uns['transcripts'] = df_temp

{cellsp-2.2 → cellsp-2.2.3}/cellSP/characterize/_sprawl.py

@@ -122,7 +122,7 @@ def run_sprawl(adata_st, methods = ['Peripheral', 'Radial', 'Punctate', 'Central
                 cell_dict['boundaries'][0], cell_dict['spot_coords'][0], cell_dict['spot_genes'][0] = vertices, spot_coords, spot_genes
                 cell_dict['zslices'].append(0)
                 cell_dict['gene_counts'] = cell.gene.value_counts().to_dict()
-                cell_dict['n'] += Z[['absX', 'absY']].values.shape[0]
+                cell_dict['n'] += cell[['absX', 'absY']].values.shape[0]
         if "gene_counts" in cell_dict:
             cell_dict['genes'] = np.unique(list(cell_dict['gene_counts'].keys()))
             if len(cell_dict["zslices"]) > 0 and min(cell_dict['gene_counts'].values()) >= 2:

{cellsp-2.2 → cellsp-2.2.3}/cellSP/datasets/_datasets.py

@@ -90,34 +90,36 @@ def st_to_anndata(transcripts, cell_type = None, cell_boundary = None):
     '''
     if Path(transcripts).suffix.lower() == ".csv":
         df = pd.read_csv(transcripts)
-        df = df[~df['gene'].str.startswith('Blank')]
-        df.sort_values(by=['uID', 'gene'], inplace=True)
-        df_st = pd.crosstab(df.uID, df.gene)
-        df_st.columns.name = None
-        df_st.index = df_st.index.astype(str)
-        if 'absZ' in df.columns:
-            result = pd.pivot_table(df, values=['absX', 'absY', 'absZ'], index=['uID'])
-        else:
-            result = pd.pivot_table(df, values=['absX', 'absY'], index=['uID'])
-        result.index = result.index.astype(str)
-        adata = ad.AnnData(X=df_st)
-        adata.obs_names = df_st.index
-        adata.obs.index = df_st.index
-        adata.var_names = df_st.columns
-        adata.obsm['spatial'] = result
-        adata.uns["transcripts"] = df
-        if cell_type != None:
-            df_ct = pd.read_csv(cell_type, index_col = 0)
-            df_ct.index = df_ct.index.astype(str)
-            adata.obs['cell_type'] = df_ct
-        if cell_boundary != None:
-            if Path(cell_boundary).suffix.lower() == ".csv":
-                df_cb = pd.read_csv(cell_boundary)
-            else:
-                raise ValueError("Incorrect file format. Only .csv files are supported.")
-            df_cb = df_cb.set_index('uID')
-            df_cb.index = df_cb.index.astype(str)
-            adata.uns['cell_boundary'] = df_cb
-        return adata
+    elif Path(transcripts).suffix.lower() == ".parquet":
+        df = pd.read_parquet(transcripts)
     else:
-        raise ValueError("Incorrect file format. Only .csv files are supported.")
+        raise ValueError("Incorrect file format. Only .csv/.parquet files are supported.")
+    df = df[~df['gene'].str.startswith('Blank')]
+    df.sort_values(by=['uID', 'gene'], inplace=True)
+    df_st = pd.crosstab(df.uID, df.gene)
+    df_st.columns.name = None
+    df_st.index = df_st.index.astype(str)
+    if 'absZ' in df.columns:
+        result = pd.pivot_table(df, values=['absX', 'absY', 'absZ'], index=['uID'])
+    else:
+        result = pd.pivot_table(df, values=['absX', 'absY'], index=['uID'])
+    result.index = result.index.astype(str)
+    adata = ad.AnnData(X=df_st)
+    adata.obs_names = df_st.index
+    adata.obs.index = df_st.index
+    adata.var_names = df_st.columns
+    adata.obsm['spatial'] = result
+    adata.uns["transcripts"] = df
+    if cell_type != None:
+        df_ct = pd.read_csv(cell_type, index_col = 0)
+        df_ct.index = df_ct.index.astype(str)
+        adata.obs['cell_type'] = df_ct
+    if cell_boundary != None:
+        if Path(cell_boundary).suffix.lower() == ".csv":
+            df_cb = pd.read_csv(cell_boundary)
+        else:
+            raise ValueError("Incorrect file format. Only .csv files are supported.")
+        df_cb = df_cb.set_index('uID')
+        df_cb.index = df_cb.index.astype(str)
+        adata.uns['cell_boundary'] = df_cb
+    return adata

{cellsp-2.2 → cellsp-2.2.3}/cellSP/visualisation/_compare.py

@@ -30,7 +30,7 @@ def apply_ztest_neg_score(c1, c2, n1, n2):
         alternative='smaller'
       )[0]
 
-def do_proportion_test(adata_st_control, adata_st_condition, file_path = None, show = True):
+def do_proportion_test(adata_st_control, adata_st_condition):
     '''
     Plot the cells in the spatial transcriptomic data.
     Arguments

{cellsp-2.2 → cellsp-2.2.3}/cellSP.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: cellSP
-Version: 2.2
+Version: 2.2.3
 Summary: cellSP.
 Author-email: Bhavay Aggarwal <baggarwal9@gatech.edu>
 License-Expression: MIT
@@ -38,6 +38,10 @@ Requires-Dist: umap-learn>=0.5.7
 Dynamic: license-file
 
 # CellSP
+[![PyPI version](https://img.shields.io/pypi/v/cellSP.svg)](https://pypi.org/project/cellSP/)
+[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](LICENSE)
+[![Docs](https://img.shields.io/badge/docs-available-brightgreen)](https://github.com/bhavaygg/cellSP#readme)
+[![Downloads](https://static.pepy.tech/badge/cellSP)](https://pepy.tech/project/cellSP)
 
 CellSP is a python package for the analysis of subcellular spatial transcriptomic data. CellSP works with datasets generated at single-modulecule resolution from technologies like Xenium, CosMx, MERSCOPE or other ISH-like data. Using existing tools [InSTAnT](https://github.com/bhavaygg/InSTAnT) and [SPRAWL](https://github.com/salzman-lab/SPRAWL/), CellSP identifies statistically signficant subcellular patterns of gene transcripts and uses a biclustering algorithm to aggregate these patterns over hundereds of cells to produce "gene-cell modules". These modules represent the consistent detection of the same subcellular pattern by a set of genes in the same cells and offer a summarized and biologically interpretable desciption of subcellular patterns. CellSP provides specialized techniques for visualizing such modules and their defining spatial patterns. Additionally, CellSP utilize Gene Ontology (GO) enrichments tests to offer functionsal insights into the genes comprising the module as CellSPll as the cells comprising the module.
 
@@ -165,11 +169,13 @@ InSTAnT identifies colocalized gene pairs based on their spatial proximity. A pr
 
 ```
 @article{aggarwal2025cellsp,
-  title={CellSP: Module discovery and visualization for subcellular spatial transcriptomics data},
+  title={CellSP enables module discovery and visualization for subcellular spatial transcriptomics data},
   author={Aggarwal, Bhavay and Sinha, Saurabh},
-  journal={bioRxiv},
-  pages={2025--01},
+  journal={Communications Biology},
+  volume={8},
+  number={1},
+  pages={1528},
   year={2025},
-  publisher={Cold Spring Harbor Laboratory}
+  publisher={Nature Publishing Group UK London}
 }
 ```

{cellsp-2.2 → cellsp-2.2.3}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
 
 [metadata]
 name = "cellSP"
-version = "2.2"
+version = "2.2.3"
 
 
 [tool.setuptools.packages]
@@ -12,7 +12,7 @@ find = {}  # Scan the project directory with the default parameters
 
 [project]
 name = "cellSP"
-version = "2.2"
+version = "2.2.3"
 authors = [
   { name="Bhavay Aggarwal", email="baggarwal9@gatech.edu" },
 ]