This repository contains source code to demultiplex four gzipped fastq files: 1 forward read, 2 forward index, reverse index, 4 reverse read, respectively. The script also need a file which specifies the valid barcodes for which to demultiplex the biological reads in the following tab-seperated format
ample group treatment index index sequence
1 2A control B1 GTAGCGTA
2 2B control A5 CGATCGAT
3 2B control C1 GATCAAGG
4 2C mbnl B9 AACAGCGA
6 2D mbnl C9 TAGCCATG
7 2E fox C3 CGGTAATC
8 2F fox B3 CTCTGGAT
10 2G both C4 TACCGGAT
11 2H both A11 CTAGCTCA
Given these, the script found in scripts/demultiplex.py has the following usage:
usage: demultiplex.py [-h] [-fq FQ [FQ ...]] [-bar BAR] [-out OUT] [-qc QC]
This file is python code for
optional arguments:
-h, --help show this help message and exit
-fq FQ [FQ ...] gzipped fastq files to de-multiplex (should be 4: 1 forward
read, 2 forward index, 3 reverse index, 4 reverse read)
-bar BAR properly formatted list of valid barcodes to demultiplex.
-out OUT output directory for which to write you files - must exist
before runtime.
-qc QC the quality cutoff for which an otherwise valid index read
mean quality score must exceed to be considered valid.
Default is 0
An example run for this might look like:
/usr/bin/time -v ./demultiplex.py -fq ../emp_files/*.gz -bar ../emp_files/indexes.txt -out ../demultiplexed_files/
This output all demultiplexed files names with R1 and R2 followed by the barcode for the resepective file. This script writes percentages and counts for each barcode to stdout like so:
Number of hopped reads = 707740
Number of undefined reads = 30783962
Number of valid reads = 331755033
Sample Barcode Total # Records Percentage of Total
GTAGCGTA 8119243 2.447
CGATCGAT 5604966 1.689
GATCAAGG 6587100 1.986
AACAGCGA 8872034 2.674
TAGCCATG 10629633 3.204
CGGTAATC 5064906 1.527
CTCTGGAT 34976387 10.543
TACCGGAT 76363857 23.018
Once the files have been demultiplexed, you can visualize which barcode hopped the most with the script found in scripts/plot_heatmap_from_hopped.py like so
/usr/bin/time -v ./plot_heatmap_from_hopped.py -fq ../demultiplexed_files/R1_hopped.fastq -bar ../emp_files/indexes.txt
to produce a plot that looks like