mothur

Cannot demultiplex fastq files with barcodes in the header

Commands in mothur
#1

Dear mothur team,

I am trying to demultiplex the seq files of a MiSeq run, which have forward.fastq, reverse.fastq and one index.fastq.

example lines from the forward.fastq:
@MISEQ:192:000000000-A88BF:1:1101:14475:1417 1:N:0:CCCTCTTTTTTC
TCCTTCTTCTCCTCTCTTTCTCCTTCTTTCTTTTTTTTCCCTTTCTCTTCTTCTTTTTTTTCCTTCCTTTTTTCCCTTTTTCTTCTCCCCCTTTCCCTTTCCTTTTCTCCTTTTTTTCTTTCTTTCCTTTTCTTTCTTCTTCCTTCCTTTTTTCTCTTTTCCCTTCTTCTCTCTCCCTCCTTCCTCCTCTTTCTCCTTCCTCTTCCTTTCCTTCCCCTTCCTCTTCTTCTCTCCCTCTTTTTTCTCCCCCC

example lines from the reverse.fastq:
@MISEQ:192:000000000-A88BF:1:1101:16318:1652 1:N:0:CACGCCATAATG
TCCTTCTTCTCCTCTCTTTCTCCTTCTTTCTTTTTTTTCCCTTTCTCTTCTTCTTTTTTTTCCTTCTTTTTTTCCCTTTTTCTTCTCCCCCTTTTCCCTTCTTTTCGTCCTTGTTTCCTTTCTTTCGCTCTATTTCTTCTTCCTTCCTTTTTTCTCTTTTCCCTTCTTCGCTCTCCCGCCGCACTCCACTTTCTCCTGCCTCTTCCTCCCCTTCCCCTTCCTCTTCTTCTCTCCCTTTTTTTTCTCCCCCC

example lines from the index.fastq:
@MISEQ:192:000000000-A88BF:1:1101:14475:1417 1:N:0:CCCTCTTTTTTC
CCCTCTTTTTTC
+
AAAAAAAAAAAA
@MISEQ:192:000000000-A88BF:1:1101:15818:1635 1:N:0:CTCCCCTCCTCT
CTCCCCTCCTCT

However, I couldn’t figure out how I should demultiplex this kind of data. Would you please help me with this?

Thank you very much for your help.

#2

Hi there -

In make.contigs you will use ffastq, rfastq, and findex (or rindex). You’ll also need to provide an oligos file to tell make.contigs which index goes with each sample. Take a look at the wiki page, see if you can figure it out. If not, come on back and let us know how far you got so we can help with the next steps.

Thanks,
Pat

#3

Dear Pat,

Thank you very much for your reply. I am trying since then. However I am having an error with the following command:

make.contigs(ffastq=fwd.1.fq, rfastq=rev.2.fq,findex=index.mid.fq, rindex=index.mid.fq, oligos=oligos.txt, bdiffs=2, pdiffs=2,processors=12)

My oligos file looks like:
forward ACACGTCTGAACTCCAGTCAC f_pr
reverse GTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT r_pr
barcode TATCGTTGACCA TATCGTTGACCA Sample1
barcode CCAGTGTATGCA CCAGTGTATGCA Sample2
barcode TGAGTCACTGGT TGAGTCACTGGT Sample3
barcode ACACCTGGTGAT ACACCTGGTGAT Sample4

A part from the error message:

[WARNING]: R_PR is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring BARCODE.
[WARNING]: TATCGTTGACCA is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring TATCGTTGACCA.
[WARNING]: Sample1 is not recognized as a valid type. Choices are forward, reverse, and barcode. Ignoring BARCODE.

Any idea?
Many thanks.

#4

Mothur does not allow reverse primers to have names. https://mothur.org/wiki/oligos_file/ Your reverse primer has a name of r_pr, which is causing the read of the oligos file to fail. Try this instead:

My oligos file looks like:
primer ACACGTCTGAACTCCAGTCAC GTCGTGTAGGGAAAGAGTGTAGATCTCGGTGGTCGCCGTATCATT desiredPrimerName
barcode TATCGTTGACCA TATCGTTGACCA Sample1
barcode CCAGTGTATGCA CCAGTGTATGCA Sample2
barcode TGAGTCACTGGT TGAGTCACTGGT Sample3
barcode ACACCTGGTGAT ACACCTGGTGAT Sample4

Also, I noticed the forward and reverse index files have the same name. The forward index file should contain the forward barcodes, and the reverse index file should contain the reverse barcodes.

make.contigs(ffastq=fwd.1.fq, rfastq=rev.2.fq,findex=index.mid.fq, rindex=index.mid.fq, oligos=oligos.txt, bdiffs=2, pdiffs=2,processors=12)