Hi everyone, I have 300 paired-end fastq files for microbiome analysis. 20 of them have empty forward files, so I can only use their reverse counterpart. I want to know if I could import those 300 fastq as single end (280 as forward and 20 as reverse).
I tried this: qiime tools import --type 'SampleData[SequencesWithQuality]' --input-path RUN/manifest-single.csv --output-path RUN/UPSTREAM/single-end-demux.qza --input-format SingleEndFastqManifestPhred33
But I get this:
An unexpected error has occurred:
Manifest for single-end reads can contain only forward or reverse reads, but not both. The following directions were observed: forward, reverse.
Is there any kind of importing method that allows what I want?
Hi @KirKara,
Technically you can just redirect your manifest file to include 280 forward plus 20 reverse files and simply name all of the directions forward and you should be able to import them in. However, I wouldn’t recommend this for a few reasons. The first is simply that those 20 reads are in a different orientation. You could try importing these 2 groups separately and merging them after but even that is not recommended. The main issue is that you are now technically dealing with reads from different regions and those are not comparable readily. Meaning those 20 samples are always going to cluster together since they will have unique ASVs not shared with the other samples even if they come from the same species. Now this may be less pronounced when you collapse those ASVs to their respective taxonomies but there will certainly still be an inherited bias with regards to different regions being used and you will likely still see artificial clustering. If this is a region with a complete overlap then you could take the reverse complement of those reads but I am assuming this isn’t the case. So the simplest option is to simply ignore those 20 samples or use the reverse reads of all your samples instead, but if those are not something you can or want to do then take a look at the fragment-insertion plugin which may be able to help you make those groups more comparable.
