Demultiplex fastq files in which barcodes and primers are already removed

Question

[kapuriakaran]

I am doing some analysis on 16S data. I have been provided with a merged fasta and a qual file from the Sequencer. There are 77 samples in this files. I converted it to fastq file. Now these file has barcodes and primers already removed from them. I want to demultiplex these files according to samples for downstream analysis. I have a mapping file with me.

I dont know how to split the file according to samples as the barcodes are already removed from them.

[Ram]

Do you not have the file(s) from before the barcode/adapter trimming? Trimming barcodes/adapters before demultiplexing makes no sense.

[Dan D]

Can you please head the converted FASTQ file so we can get a basic overview of its structure?

[Brian Bushnell]

+1. Sometimes the barcodes are in the read name.

[Dan D]

Assuming this is the case (which hopefully it is), which tool will help him do what he needs? Fill in the blank:

BB_________

[Brian Bushnell]

That would be...

demuxbyname.sh in=reads.fq out=sample_%.fq prefixmode=f length=8

...assuming the bar code is the last 8 characters of the read name. Alternately, if you have a list of barcodes, you could use "substringmode=t names=barcodes.txt".