I am trying to understand the UPARSE pipeline developed by Dr. Edgar. In his manual, he suggested to do sample pooling. However, in the example commands he showed on the website, he did not include this step. If you have used this procedure, did you follow his suggestion?
The point is if you're interested in comparing communities from different samples that were sequenced in different runs (or if they have individual files for some reason), they need to be combined in order for the OTUs to be called similarly. Say you're comparing a before and after scenario, you what to know that OTU #1 which is decreases in abundance in the after samples is the same OTU #1 as in the before samples because OTU calling is somewhat random. Hope I made myself clear. Anyway the way to do it is as you mentioned is to combine the files. You can do it using the cat command in linux/macOS or the copy /b command in windows. If you have samples from different sequencing runs that use the same bar-codes you should marge the files only after demultiplexing (i.e. after bar-codes have been removed and you generated fasta files) but before you de-replicate. If you don't have samples that use similar bar-codes you can marge the fastq files but I suggest using mothur for that.
Hope that's clear
Ido
- Badges
- Science method