COMIRI

Please note: this tool is not finished. This README will be updated as the tool is developped.

Introduction:

COMIRI (COsensus Metacaller for Immune Receptor Identification) automates the extraction of T-cell and B-cell CDR3 sequences from bulk RNA sequencing. The program can run the following algorithms:

• CATT
• MiXCR
• TRUST3
• TRUST4
• V'DJer

Overview:

Features:

• Single command to run all tools and dependencies, with adjustable parameters for each tool, filtering step and consensus method
• Possibility of running with Docker or Singularity containers, for increased portability and reproducibility
• Parallelization with Parsl, a flexible and scalable parallel programming library for Python. Parsl allows for a modular allocation of resources to run each step concurrently, while respecting any dependencies between programs

Installation:

pip install comiri

Usage:

comiri -f FASTQ_PATH -o OUTPUT_DIR <OPTIONS>

-f, --fastq_path FASTQ_PATH Path of the input fastq files
-o, --output_dir OUTPUT_DIR Path of the output directory

Options:
-h, --help Show this help message and exit
-g, --genome_version {hg19,hg38} Genome version to use
-p, --run_program {TRUST3,TRUST4,MiXCR,VDJer,CATT} List of programs to run
-m, --run_mode {docker,singularity,local} Run mode for pipeline
-r, --receptor {TCR,BCR-heavy,BCR-light} List of receptors to extract
-c, --config CONFIG Config file to use for running Parsl
-j, --json JSON JSON file to use for running Parsl
-s, --single_end Run in single-end mode

Poster:

CSHL Biological Data Science Conference (2020)