In one project using sets of n fusion primers for multiple sets of n samples

Dear Community and Qiime development team,

I have an expensive set of 96 fusion primers which target the 18S gene in mixed template DNA extracts. The design of the primers (i.e. specifically non-target binding sequences) follows the EMP pages (at http://press.igsb.anl.gov/earthmicrobiome/protocols-and-standards/18s/ ).

I have already used those primers on 96 samples of one particular study, and would like to re-use these primers - in order to save money - in a second set of samples belonging to the same study. I would prefer not doing this, but of course need to work cost efficiently.

During sequence pre-processing I remove barcodes using qiime cutadapt trim-paired and later, to remove remnants, a second time using cutadapt ( the representative set of sequences is exported to fasta, re trimmed, and re-imported). Despite this, I am sure that some barcode remnants remain, and of course the barcode and tag sequences are contained in the metadata of all samples.

Will duplicated barcode sequences in the metadata prevent me from processing multiple independent sets samples using the same primer pairs? Or to put differently, is it possible in one project to use sets of n fusion primers for multiple sets of n samples, without collisions during later analysis?

I would be looking forward to hearing from you in due course.

Kind regards,

Paul

Hi @Paul_Czechowski ,

In a lot of my projects, we reuse the same set of barcodes over and over again. (The American Gut is probably an extreme example, but there are definitely not 10,000+ unique barcode sequences.) The best way to approach the problem is to keep your sequencing sets (batches of barcodes) separate until you’ve finished demultiplexing, since demultiplexing is where you “name” the barcodes, and match the barcodes back to the sample name. However, you can’t combine the samples on the same sequencing run, for example.

If you’re using a denosing/closed reference picking method, you could even cluster/denoise on the seperate batch.

The one thing I’d as a consideration, though, is the relative cost of buying additional primers that let you run 192 samples on your current/future sequencing runs verse the cost per sample for sequencing if you run 96 samples per run.

Best,
Justine

1 Like

Hello Justine,

thank you very much I appreciate the response.

Kind regards,

Paul

This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.