Atlas-level data integration of BAL samples from severe COVID-19

This repository contains code for the processing and analysis of scRNA-seq samples from patients with severe COVID-19.

Directory Structure

• bin/ - Scripts used for the analysis
• docs/ - Document files
• data/ - Data output (large, not included in git)
• workflows/ - Workflows executing scripts in bin/
• reports/ - Reports for exploratory analysis
• analysis/ - Analysis output (not included in git)
• man/ - Manuals explaining the rationale behind analyses
• envs/ - conda environment YAML files
• LICENSE - The project license

Data Availability

Shared on web portal: TODO

Download as part of workflows, described in bin/dataset_*

Analysis workflows

The analysis workflow include download & setup of datasets. Quality control, data integration, cell annotation, evaluation using (scIB) metrics and visualization of reports.

The workflows are written in bash and can be submitted using SLURM.

To reproduce the analysis please

1. Install conda (follow instructions and accept defaults), close and re-open the console

   curl -o miniconda.sh https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
   bash miniconda.sh
   rm miniconda.sh
   

2. Install mamba

   conda install -c bioconda mamba -y
   mamba init
   

3. Clone and enter the git repository (if you want to specify the directory replace '~' with your local path)

   cd ~
   git clone https://github.com/saliba-lab/covid19-bal-atlas-integration.git
   cd covid19-bal-atlas-integration
   

4. Create conda environments

   mamba env create -f envs/R.yml
   conda env create -f envs/integration.yml
   conda env create -f envs/scib.yml
   

5. Execute workflows (beginning with the count matrix)

   bash workflows/setup_BCB.sh
   

Addendum

To use nextflow workflows please install java

sudo apt install default-jre