search

jfjlaros / demultiplex

Versatile FASTA/FASTQ demultiplexer.
MIT License
32 stars 5 forks source link

Demultiplexing not working? #4

Closed ViriatoII closed 4 years ago

ViriatoII commented 4 years ago

Hey, Maybe I got something wrong, but it's not demultiplexing my fasta file. I have a big file of pacbio reads, with Pacbio primers + individual barcode. I formated it as follows in a barcodes.csv:

D408 GCAGTCGAACATGTAGCTGACTCAGGTCACCACATATCAGAGTGCGGGTAGT
FAP1360A GCAGTCGAACATGTAGCTGACTCAGGTCACACACACAGACTGTGAGGGTAGT
Ky226 GCAGTCGAACATGTAGCTGACTCAGGTCACACACATCTCGTGAGAGGGTAGT
P092 GCAGTCGAACATGTAGCTGACTCAGGTCACCACGCACACACGCGCGGGTAGT
W64A GCAGTCGAACATGTAGCTGACTCAGGTCACCACTCGACTCTCGCGTGGTAGT

(I erased the reverses, but am not sure if your program automatically reverses barcodes) Then run: demultiplex demux barcodes.csv myinput.fasta -r

I only get one input: myinput_UNKNOWN.fasta with all the reads. Whats wrong? I've also trying running it with only one barcode in the .csv file. Same problem.

Cheers,

jfjlaros commented 4 years ago

The command you use assumes that the barcode is at a fixed position in the read, the location of which can be specified using the -s and -e options.

I assume though, that your barcodes are not at a fixed position, for these cases the match command can be used. This command indeed also checks the reverse complement.

ViriatoII commented 4 years ago

It works! Thank you for the swift reply!

By the way, are the barcodes also trimmed from the demultiplexed sequences? At a first sight it appears not.

jfjlaros commented 4 years ago

No, the reads remain unchanged.

ViriatoII commented 4 years ago

Thanks! Have a nice day.