#count #read #index #stats #identify #indices #distribution

bin+lib bcl2fq-stats

bcl2fq-stats is designed to give a quick overview of read count distribution over the given indices, and identify potential index mismatches from the undetermined read counts

2 releases

0.1.1 Apr 15, 2023
0.1.0 Apr 15, 2023

#735 in Science

Apache-2.0

31KB
318 lines

bcl2fq-stats

CIcrates.io

Illumina sequencing data is always demultiplexed with Bcl2fastq, but it's not always easy to get a table of read counts from the demultiplexed data. bcl2fq-stats is designed to give a quick overview of read count distribution over the given indices, and identify potential index mismatches from the undetermined read counts.

Usage

The program takes Stats.json in the bcl2fastq output folder as input:

bcl2fq-stats --json-file data/Stats.json

Installation

git clone https://github.com/wckdouglas/bcl2fq-stats.git
cd bcl2fq-stats
cargo install --path .

or using docker:

docker pull ghcr.io/wckdouglas/bcl2fq-stats:main

Dependencies

~4.5–7MB
~120K SLoC