gcell is a comprehensive toolkit for genomic data analysis, focusing on cell type-specific regulatory analysis, DNA sequence manipulation, protein structure prediction, and pathway analysis. It integrates various modules to facilitate the study of different aspects of gene expression regulation.
- I dropped
graphvizandpygraphvizdependency due to complexity of maintain installing across different platform. It only affects the network layout for drawing causal graph.nx.spring_layout(G)is used instead ofnx.nx_agraph.graphviz_layout(G)now, which is uglier unforturnately. - Feature: Now you can load pre-infered cell types on getdemo website easily:
from gcell.cell.celltype import GETDemoLoader
g = GETDemoLoader()
print(g.available_celltypes) # this gives you a list of cell type names
g.load_celltype('Plasma Cell')- Fix:
zarrhas been limited to<3.0.0to avoid s3 problem
The long term goal of this package is to create a open-source, community-involved python-centric playground/tool-set for future AI Agent to discover new biology through predictive model.
- Celltype: Tools for cell type analysis, including gene expression and motif analysis. Basis for
get_modelinterpretation analysis. - DNA: Functions for DNA sequence manipulation, motif scanning, and track visualization.
- RNA: Classes for handling GENCODE gene annotations and GTF files.
- Protein: Functionality for protein domain analysis (Uniprot, InterPro) and AlphaFold2 predictions parsing as well as retrieve protein-protein interaction networks from the STRING database..
- Pathway: Tools for pathway (GO, KEGG, Reactome, etc.) analysis using gprofiler.
pip install git+https://github.com/GET-Foundation/gcell.git@main
gcell is open-source software licensed under the MIT License. See the LICENSE file for more details.