Cell type Identification using Transcription factor Analysis and Chromatin accessibility
A pipeline for single-cell multiome (scATAC + scRNA) data that identifies cell types from transcription-factor motif activity. RNA is used only to derive labels; the classifier itself is trained purely on TF activity, so the model learns chromatin-level regulatory signal.
| Stage | Steps | Tools |
|---|---|---|
| 1. Preprocessing (R) | multiome H5 → joint RNA+ATAC QC on shared barcodes | Seurat, Signac |
| SingleR labels from the Monaco immune reference | SingleR, celldex | |
| JASPAR motif scan → chromVAR per-cell TF activity | JASPAR2020, motifmatchr, chromVAR | |
| 2. Machine learning (Python) | class composition, unsupervised feature cleaning | pandas, scikit-learn |
| Logistic Regression, Random Forest, SVM, XGBoost | scikit-learn, xgboost, imbalanced-learn | |
| TF ↔ cell-type association (Mann-Whitney + BH-FDR + effect size) | scipy | |
| figures, tables and a JSON report | matplotlib, seaborn, networkx |
Scope. cellitac has been developed and tested on human PBMC multiome data, using the Monaco immune reference for SingleR labels. It works on any human tissue whose cell types are covered by that reference (blood, bone marrow, immune infiltrates).
Genome builds: hg38 (default) and hg19.
- Linux or macOS (Windows via WSL)
- Python 3.9 – 3.12
- Conda / Miniconda
- R ≥ 4.3 with the Bioconductor packages listed below (for the preprocessing stage). The ML stage runs on Python alone.
Take the R packages from conda as pre-built binaries — do not let BiocManager compile them from source.
conda create -n cellitac -c conda-forge -c bioconda -y \
python=3.11 rpy2 r-base=4.4 \
r-seurat r-signac r-data.table \
bioconductor-jaspar2020 bioconductor-tfbstools \
bioconductor-motifmatchr bioconductor-chromvar \
bioconductor-singler bioconductor-celldex \
bioconductor-biovizbase bioconductor-rtracklayer \
bioconductor-summarizedexperiment bioconductor-biocparallel \
bioconductor-bsgenome.hsapiens.ucsc.hg38 \
bioconductor-ensdb.hsapiens.v86
conda activate cellitac
pip install cellitac