CoCnT Analysis Scripts

This repository collects the analysis scripts used to process and analyze the PBMC CoCnT dataset and T cell depleted bone marrow CoCnT dataset.

The code is organized by analysis workflow directory: each workflow directory contains scripts for one part of preprocessing, donor deconvolution, ArchR analysis, or pseudo-bulk track generation.

Detailed instructions for each stage are documented in the README.md file within the corresponding subdirectory.

What's not included in this repo:

About sequencing data demultiplexing and mapping, please refering to our Nextflow workflow: https://github.com/vari-bbc/sciCT_pipeline

It needs BAM files to use SouporCells to identify the cell doublets. But the BAM files are very big for shipping with this repo or in Zenodo. So the souporcell step will only provide the script and the final output. If you are using scCoCnT and want to run SouporCells, please check out our Nextflow workflow here: https://github.com/vari-bbc/souper-star

Repository Layout

• run_PBMC_coCnT_analysis/ Analysis scripts for the PBMC CoCnT dataset.
• run_BM_create_arrow/ Generate ArchR Arrow files from mono-occupancy and co-occupancy BED inputs.
• run_BM_souporcell/ Run Souporcell for donor deconvolution and doublet identification.
• run_BM_archR_analysis/ Perform the core ArchR-based analysis.
• run_BM_bed_profile/ Generate fragment QC summaries and pseudo-bulk BED/BigWig tracks.
• run_BM_motif Perform motif enrichment analysis and TF footprinting.
• lib Helper functions sourced by multiple analysis scripts.
• data/ The fold containing the raw input, it is not included in this repo.

Prepare the data folder

Please download the data folder from Zenodo, which contains the raw Bed files and the sample sheet for the analysis. BED files and sample sheet for Sun & Baird et al. CoCUT&Tag maps linked chromatin states at single-molecule, single-cell resolution

Also for the reference scRNASeq data, please download from here: https://figshare.com/articles/dataset/Expression_of_97_surface_markers_and_RNA_transcriptome_wide_in_13165_cells_from_a_healthy_young_bone_marrow_donor/13397987?file=41038073

And place the file under the data folder.

Also for the reference genome files, please download the hg38 reference genome files and the genome size files, and place them under the data/ref_genome folder.

They should named as hg38.sizes, hg38_gencode.fa.