How to split a FASTQ file with ITS and 16S reads in it in R

Question

[katie.miaow]

Hi,

For my last batch of sequencing I multiplexed my 16S and ITS sequencing PCRs on to one run, and as a result each forwards and reverse FASTQ file contains a mixture of 16S and ITS sequences. I would like to separate them out (I guess by searching through for each primer sequence and using that to categorize the reads), so I can continue down my usual pipeline with 16S and ITS separately (which is R with Dada2, DESeq2 and Phyloseq, with cut-adapt to remove primers). All the tutorials I can see assume only one gene sequence in the FASTQ file. I only have basic R skills and can't easily find a solution to this problem, does anyone have any ideas on the best way to go about this?

Thank you

Answer 1

[Mark]

Use Rsubread package to align the reads to whatever reference makes sense. The align function will output both aligned and unaligned reads.

edit: You multiplexed your PCR and sequenced this mixture, am I understanding correctly? Or are you after an R based method for demultiplexing FASTQ files that were multiplexed using illumina tags?

Answer 2

[katie.miaow]

Hi I multiplexed 16S and ITS PCR for all my samples. I have FASTQ files already de-multiplexed by the indexing primer tags, so I know what the sequence is for each sample but the files contain both 16S and ITS reads. I was wondering about essentially de-multiplexing them again based on the 16S and ITS primers but as they come to me after this step I'm unsure how to do this myself. I was considering just putting them down the pathway together and they would separate on alignment but given Dada learns error rates ect and looks for chimeras I felt it would be more elegant to try to separate the files if possible as the sequences I imagine would have differences in those parameters. Thanks for your time so far!

[Mark]

Please reply to my post instead of posting an "answer". New posts are for answers only.

In that case, you should align to a 16S reference to filter.

[katie.miaow]

Thanks for your thoughts! I'll give this a try.