Add the ability to reverse complement sequences before writing them

[ElDeveloper]

Improvement Description
split_libraries_fastq.py provided a --rev_comp flag, that would let you reverse complement the sequences before they were written out.

After talking with @wasade, we both agree that such functionality might not be appropriate for q2-quality-filter, but we couldn't really figure out where this would fit better, so also opening this issue hoping this can get relocated to the proper repo.

[jairideout]

q2-demux?

[ElDeveloper]

Maybe, though then there's no way to reverse complement per-sample FASTQ files.

[jairideout]

Are you wanting support for reverse complementing any kind of sequence data stored in .qza files, not just during demultiplexing?

[ElDeveloper]

Yes. This was the case in QIIME1, since the functionality was available in split_libraries_fastq.py and multiple_split_libraries_fastq.py used split_libraries_fastq.py.

[jairideout]

Sounds good! Are you available to work on this for the 2017.9 release?

[ElDeveloper]

Yes, where should this be added to?

[gregcaporaso]

What's the use case for reverse complementing fastq files? I worry that that might lead to assumptions about the quality profiles being violated (e.g., median quality will be positively correlated with sequence position in the resulting files, rather than negatively correlated, which could mess things up downstream). It seems like a reverse complement action might make more sense if it operated on FeatureData[Sequence] instead.

[ElDeveloper]

@gregcaporaso, after solving the problem I had with that particular dataset, I don't think I have a use-case anymore. Seems like the only place where this could matter would be with the FeatureData[Sequence] as you mention. I am down to put that in place if you all think this still makes sense.

[ebolyen]

Since it looks like this isn't immediately necessary, we're dropping it from our sprint plan. We'll keep an eye out for more situations where this could be needed.

[ElDeveloper]

Sounds good to me.

…

On (Sep-13-17|18:40), Evan Bolyen wrote: Since it looks like this isn't immediately necessary, we're dropping it from our sprint plan. We'll keep an eye out for more situations where this could be needed. -- You are receiving this because you authored the thread. Reply to this email directly or view it on GitHub: #33 (comment)

[ARW-UBT]

Would the following be a use case?
We continously analyse fungal ITS amplicons (paired-end data). Due to the length heterogeneitiy of ITS, it is not possible to joine forward and reverse reads for many of them. Therefore, joined reads cannot be processed by
a) q2 quality-filter q-score-joined (followed by q2 deblur denoise-other) or by
b) q2 dada2 denoise-paired.
I such situations, one can process single end data from R1 reads (forward), but this fails for R2 reads, probably because the 'antisense' R2 sequences are sorted out at q2 deblur denoise-other.

Since independent analyses can be combined in q2, I thought about

1. reverse-complement the FeatureData[sequence] artifact
2. analyse the rc-R2-sequences as single-end data and
3. finally combine both analysis results...

Best regards