How to demultiplex but not merge reads

[lina....@gmail.com]

Hi all,

I have data from a miseq run in three gzipped fastq files: R1, R2, and the index file (I1). I also have a mapping file with barcode sequences.

I need to demultiplex these files and am trying to figure out how to do this without merging reads. In the end, I need one R1 and one R2 file with the fastq sequences for each sample.

I was looking at split_libraries_fastq.py:

Demultiplex and quality filter (at Phred >= Q20) two lanes of Illumina fastq data and write results to ./slout_q20.:

split_libraries_fastq.py -i lane1_read1.fastq.gz,lane2_read1.fastq.gz -b lane1_barcode.fastq.gz,lane2_barcode.fastq.gz --rev_comp_mapping_barcodes -o slout_q20/ -m map.txt,map.txt --store_qual_scores -q 19

But I don't have two mapping files, so I'm assuming this might not be the best approach.

Thanks for any advice you might have!

Cheers,

~Lina

[justink]

I think that split_libraries_fastq.py will work if you do one read at a time, as in this example:

Demultiplex and quality filter (at Phred >= Q20) one lane of Illumina fastq data and write results to ./slout_q20. Store trimmed quality scores in addition to sequence data.:

 split_libraries_fastq.py -i lane1_read1.fastq.gz -b lane1_barcode.fastq.gz --rev_comp_mapping_barcodes -o slout_q20/ -m map.txt --store_qual_scores -q 19

hopefull that gets what you're looking for.

-Justin