TAC-seq data analysis

This repository contains the sequencing data analysis software for TAC-seq.

Requirements

• Linux-based OS
• FASTX-Toolkit
• Git

Setup

Use the following commands to setup TAC-seq data analysis software on terminal:

1. Install FASTX-Toolkit
2. Install Git
3. Download the analysis software using Git: git clone https://github.com/hindrek/TAC-seq-data-analysis
4. Navigate to analysis location: cd TAC-seq-data-analysis
5. Make tacseq executable: chmod +x tacseq

Usage

tacseq [options] <command>

Analyze TAC-seq data.

options:

• -h display help and exit

commands:

• prep prepare samples (FASTQ files) for counting
• count count reads and molecules per sample and target

tacseq prep [options]

Prepare samples (FASTQ files) for counting.

mandatory:

• -i input file: gzip compressed/uncompressed FASTQ file or '-' as standard input (stdin)
• -t target file: target file format is based on FASTX Barcode Splitter barcode file format
• -o output directory

optional:

• -h display help and exit
• -m mismatches: number of allowed mismatches per target sequence (default: 5)

tacseq count [options]

Count reads and molecules per sample and target.

mandatory:

• -i input directory: tacseq prep output directory

optional:

• -h display help and exit
• -u UMI threshold (default: 2)

Target file format

Target file is a text file which contains a list of targets. Each line has to contain a target ID (must be alphanumeric) which is followed by the target sequence (only A, C, G and T characters are allowed). Target ID and sequence are separated by a TAB character.

Target file example:

TARGET1 TAGGATAGGTGGATTCGGGAACTCCCCGATAGTTTTGTCACATCGACATACTAA
TARGET2 CCAAAGCTTCAACGGACATAGTGTACATACCTACCGTGTTTCCCAGCACCTTCC
TARGET3 CTGCTGTTGCCGCCTGGGGTTTACGCGTGTTGGAGATTGAGTAGCCTCCTCGGC

Output

tacseq prep outputs a directory for each sample with:

• 3 sub-directories with files for each target:
  • loci
  • umis
  • merged
• 2 intermediate files:
  • trimmed.fasta
  • umi_joined.fasta

tacseq count outputs read and molecule counts per target for each sample.

Run example

• Step 1 - prepare samples: ./tacseq prep -i example/samples/sample1/sample1.fastq.gz -t example/targets.txt -o example/output/sample1/ -m 5
• Step 2 - count molecules and write results to counts.tsv file: ./tacseq count -i output/sample1/ -u 2 > counts.tsv