Advice on combining sff's

Theory behind mothur
1

Hello,

I had a quick question to ask some of you more experienced analyzers out there.

We are going to be doing some sequencing in the near future. We were throwing around the idea of splitting our samples up into a few different wells, and sequencing them in batches, so that we could reduce the number of barcodes we were using. My question is, if we do this, with multiple individuals getting the same barcode, is it still possible to do the analysis without the program thinking that two sample sets are one in the same? I did see the sff.multiple command, but I didn’t really see any info relevant to the barcode issue when I glanced through.

I was thinking that you could use a word processor to find and replace the duplicate barcode with a different, alternate one? Or is there a command in mothur that would fix this easier?

I was not sure about this, as in the past, our sequencing has been returned in a single sff file, and this is uncharted territories for our team.

Any help is appreciated!
Thanks!

2

Hi,

Yea, if you look at the SOP there’s a step where you use the trim.seqs command just after denoising. At this step, the barcodes are removed and you just have the bacterial (or whatever) sequence. At that stage, you’re free to pool data from different runs without fear of accidentally merging barcodes together. This is all the sff.multiple command does anyway, it runs the SOP up to this stage for each input file, then merges them all.