I have some confusion. I have to sequence files. Both are pair-end reads. But one of them has a very poor reverse read quality. So I decided to merge pair ends for the good one and then denoise it by deblur at 250 sequence length. And for the poor one (low-quality reverse read) I have decided to work with only the forward read. And trimmed at 101 sequence length (as all the sample remains if I cut at 101 length).
So now my question is, is it okay to do the above steps and then merge two table.qza and rep-seqs.qza files for further analysis together?
Hi @turtle
I would rather say that it is not OK to merge the outputs of so differently preprocessed sequences.
In similar situation I would either discard the second dataset at all or process only forward reads from both datasets with identical parameters and then merge tables and sequences. Which option to choose depends on your priorities.
If I work with only forward reads, would you suggest me to trim the 1st one at 101 lengths and the second one at 215 lengths and then merge them? Or do I just have to trim both of them at 101 lengths by deblur and then merge them?
Can you please tell me how can I merge two datasets that I downloaded from the SRA database with the help of qiime2? I mean I want to merge the datasets before deblur denoising steps. It would be very helpful if you could provide me the code or the link from where I can get the instructions.
I am not aware how to perform this task since I never tried it. You can create a new topic (subject changed from originally posted) to see if it can be accomplished in Qiime2, or export both datasets and then reimport as one artifact.
