ATG_fastq_demultiplexer

This script uses FASTX Toolkit with default parameters to split a fastq file into multiple fastq files based on the sequences at the begining of line (bol) and end of line (eol). Merge the paired-end fastq files before using this script (alternatively use ATG_fastq_merger, which is a python script that uses FLASH for merging illumina NGS R1 and R2).

Requirements

• fastx_toolkit package: can be installed via conda, note: fastx_barcode_splitter.pl is expected to be in the path after the installation of fastx toolkit.
• python3
• Tested on a linux machine, no reason to not to work on Mac and Windows.

Usage

python atg_fastq_demultiplexer.py [--fastq-dir PATH_TO_FASTQ_DIR] [--barcodes-file PATH_TO_BARCODES_FILE] [--out-dir PATH_TO_OUT_DIR]

or

python atg_fastq_demultiplexer.py [-i PATH_TO_FASTQ_DIR] [-f PATH_TO_BARCODES_FILE] [-o PATH_TO_OUT_DIR]

Prepare a tab delimited barcodes_info.tsv file with the following four column names: sample_name, barcode_bol, barcode_eol, and demuxed_name. For example:

sample_name	barcode_bol	barcode_eol	demuxed_name
mysample_1	CAGTT	AACTG	mysample_1_CAGTT_AACTG_TRT_x_Rep_1
mysample_1	CAGTT	TCTCT	mysample_1_CAGTT_TCTCT_TRT_x_Rep_2
mysample_1	CAGTT	GATCA	mysample_1_CAGTT_GATCA_TRT_x_Rep_3
mysample_1	CAGTT	TTCGG	mysample_1_CAGTT_TTCGG_TRT_y_Rep_1
mysample_1	AGAGA	AACTG	mysample_1_AGAGA_AACTG_TRT_y_Rep_2
mysample_1	AGAGA	TCTCT	mysample_1_AGAGA_TCTCT_TRT_y_Rep_3
mysample_1	AGAGA	GATCA	mysample_1_AGAGA_GATCA_TRT_z_Rep_1
mysample_1	AGAGA	TTCGG	mysample_1_AGAGA_TTCGG_TRT_z_Rep_2
mysample_1	TGATC	AACTG	mysample_1_TGATC_AACTG_TRT_z_Rep_3

bol = begining of line (i.e., read), eol = end of line (i.e., read).

Important Note 1: sample_name must be alphanumeric (i.e., avoid &, $, @, -, %, * and empty space in the file names).

Important Note 2: Fastq files in the FASTQ_DIR are expected to have .fastq suffix and their names match the "sample_name" in BARCODES_FILE.

Outputs

• out_dir/demux_name.fastq contains the barcode splitted fastq file(s). Note that barcodes from the begining and of reads are not trimmed.
• out_dir/fastx_splitter.log contains the fastx statistics output (e.g., number of reads per splitted files and number of unmatched reads).

References

• Fastx toolkit: http://hannonlab.cshl.edu/fastx_toolkit/commandline.html.

Bugs

Please report errors/bugs to: Amir.Taheri.Ghahfarokhi@gmail.com