This single cell ATAC-Seq analysis pipeline is designed for advanced analysis of dataset, produced by 10X Genomics Cell Ranger ATAC. Aggregated datasets are also supported!
In addition to 10x Genomics results it offers:
- Capable to process aggregated data by 10X Genomics Cell Ranger ATAC.
- Summary on different conditions in case of aggregated dataset
- Flexible data processing with t-SNE/UMAP visualizations in low dimensions space
- User defined markers visualization as a heatmap
- Closest genes annotations for peaks and clusters
- Annotated markers analysis
- Bigwig and BED files for clusters and markers ready-to-be-visualized in JBR Genome Browser
- Data preparation for Single Cell Explorer
- Save all the figures to ready for publication PDF format
- Launch batch cell ranger processing.
NOTE: we don't launch it in parallel because of martian framework used by Cell Ranger ATAC.
for SAMPLE in $(ls *.fastq.gz | sed -E 's/_S[0-9]_L00.*//g' | sort --unique); do
cellranger-atac count --id=cra_${SAMPLE} --fastqs=${WORK_DIR} --sample ${SAMPLE} --reference ${REFERENCE}
done
- Create aggregation file
merged.csv
library_id,fragments,cells
<id1>,<path1>/outs/fragments.tsv.gz,<path1>/outs/singlecell.csv
...
<idN>,<pathN>/outs/fragments.tsv.gz,<pathN>/outs/singlecell.csv
- Launch aggregation
cellranger-atac aggr --id=<id> --csv merged.csv --normalize=depth --reference=${REFERENCE}
Conda environment sc-atac-explorer can be easily created for launching Jupyter Notebook:
```
conda env create -f environment.yml
conda activate sc-atac-explorer
```
Launch jupyter notebook to proceed with the pipeline.
```
conda activate sc-atac-explorer
jupyter notebook
```