Interested in Illumina myseq 16s barcoding

[giderk]

I've just signed up this forum and am introducing myself as suggested. I am most curious in 16s rRNA barcoding using the protocol that Illumina published; 16s Metagenomic Sequencing Library Preparation.

If anyone has experience following this protocol, I'm eager to speak with you. My interest is in adapting it to 18s.

Also, I didn't readily see a "search" forum form or field on the home page of this website. Does one exist?

[cliffbeall]

We have modified that protocol to work on different parts of the 16S gene and also other custom amplicons. All you do is change the gene-specific parts of the primers for the first round of PCR. The second round primers, which add the barcodes (indexes in Illumina parlance) remain the same.

It works ok, though for some reason we see more non-specific amplification and low quality than we were used to from 454, so we have been throwing out a lot. Since there is so much data to begin with, though, it doesn't seem like a huge problem.

Edit: I just realized that you were referring to the entire process of 16S rRNA sequencing as "barcoding". I would probably avoid that language because it would lead to confusions like mine, with the multiplex indexing process. At any rate, my advice is the same.

[giderk]

Thank you. My main obstacle at the moment is selecting primers for this protocol that would give me the best resolution for species-specific identification. Others on this forum have suggested a target region that is much less than 550bp (maximum suggested by illumina for this protocol). It sounds like the reads get ugly towards the end, so 50bp of overlap usually isn't enough. Is this your experience, as well?

[cliffbeall]

Thank you. My main obstacle at the moment is selecting primers for this protocol that would give me the best resolution for species-specific identification. Others on this forum have suggested a target region that is much less than 550bp (maximum suggested by illumina for this protocol). It sounds like the reads get ugly towards the end, so 50bp of overlap usually isn't enough. Is this your experience, as well?

Yes, that's our experience. We have done 500, but with a curated database and a phyllotype approach plus as I mentioned throwing out a fair amount of data.