This script will demultiplex reads according to a barcode file. It will output interleaved files (zipped), non interleaved files (I think not zippped), trimmed files (adaptor contamination removed) and merged files (if the reads overlap, they will get merged and modify the headers to be qiime ready). It automatically detects the illumina nomenclature for index files. Samples are demultiplexed 96 at a time by default. The script will loop until all samples are processed.
- Anthony Bolger's
trimmomatic.jar - FLASH, a tool for merging paired-end reads
- [FLASH2, improved version of FLASH] (https://github.com/dstreett/FLASH2)
perl demultiplex_dualBC.pl <options> <illumina_directory> <mapping_file> <output_directory> <filename_core>`illumina_directory The directory where the raw Illumina output lives.
mapping_file A tab delimited, QIIME-style mapping file, like so :
#SampleID BarcodeSequence ReverseBarcode
SAMPLE1 AACGCTAA GACTGGTT
SAMPLE2 AACGCTAA TTAGCGTT
SAMPLE3 AACGCTAA GTAGTCTT
output_directory The name of the directory where output will be written.
filename_core Some sort of identifier added to the files before the sample
names, which will look like <core>_<samplename>.faa.
--chunk-size : Will process this many samples at a time. This is to prevent having to split the mapping file in case there are lots of samples. 96 seemed like a good/safe number. Increase at your own risk.
--trim-file : Adapter sequences that need to be removed from the raw data in
the case of contamination. An example is found in adapter_all_16s.txt.
--trim-tool The path for the trimming tool that will be used for removing
adaptor contamination. The file it is looking for is Anthony Bolger's
trimmomatic.jar, which can be obtained from the Usadel Lab software
page.
--reverse For obscure reasons, this option is always required for demultiplexing 16S data.
--q [n] The minimum quality score that will be kept during the QC. 20 is the default.
--read-len : Read length. 250 or 300 these days.
--frag-len Length of the fragment you are targeting. For standard 16S you will want something around 253.
--min-overlap This will be used to merge the reads. Won't merge reads that overlap less than 10 residues (default 10).
--max-overlap This will be used to merge the reads. It won't merge more than 70 residues. This should be adjusted depending on fragment length Vs read length. I use 120 often for this. Default is 70.
--mismatch-ratio Do not merge reads if the overlapped residues have a mismatch ratio over this number. Default 0.25
--skip-merge Use this flag if you do not want to merge the reads.
--phred Phred alphabet used. 33 is default and what recent sequencing machines use.
--skip_interleaved [1|0] Set this flag if you want to skip over the interleave file generation.
--no-mismatch Skips over the mismatch gadgetery added in this script. By default it will allow 1 mismatch per barcode half. Barcodes are set by 2 index reads. It allows 1 mismatch for each index read.
--min-read-length [n] Discards reads that end up being shorter than the number after the QC (and maybe merging).
--demul-only [n] Skips all QC and merging steps