Split rep-seq by sample id

[sarah.lital]

Hi, Is there a way to generate sample specific fasta files? I know that qiime 1 had this option using split_sequence_file_on_sample_ids.py but I haven’t found a qiime2 counterpart. I am using version 2019.4.

Thank you for the assistance.

[andrewsanchez]

Hi, @sarah.lital and welcome to the forum!

Without knowing what you are trying to accomplish, I suspect filtering actions will help you out. Both the demux and the feature-table plugins have a filter-samples action, e.g., qiime demux filter-samples and qiime feature-table filter-samples. There are different ways to filter data using qiime 2 and your goals will dictate how you do it. Please see the Filtering tutorial and the section on identifier-based filtering in particular.

I hope that helps!

Cheers,
Andrew

[sarah.lital]

Thanks for the reply! I’ve checked out the filtering options and it doesn’t seem to do what I’m trying to accomplish. I have data comprised of 5 different miseq runs which I’ve merged, for a total for 475 samples and 8000 OTUs. I have the rep-seq artifact from DADA2 as well as the frequency table, which includes all 8000 OTUs and I would like to generate a separate sequence file for each of the 475 samples based on each OTU’s presence in the sample.

I hope this is clearer, thanks very much for the help!

[andrewsanchez]

Hi, @sarah.lital,

I should have added that qiime tools export can be used to export the resulting artifacts from filtering actions. Please see the Exporting data tutorial for examples.

For example:

➜ qiime tools export --input-path seqs.qza --output-path seqs-exported.fna
Exported seqs.qza as QIIME1DemuxDirFmt to directory seqs-exported
➜ ls seqs-exported
seqs.fna

Let me know if qiime tools export does the trick for you.

Happy qiime-ing!
Andrew