GBS: eliminating adpater contamination in library for Illumina platform

[anjama]

We're in the process of trying to create our own GBS libraries for various species. The step we're having the most trouble with right now is optimizing the adapter concentration to avoid having adapter dimer in the final library. My understanding is that Illumina recommends that adapter contamination account for at most 0.5% of the library, and the protocol calls for even less. We aren't entirely clear on the appropriate way to calculate this. The bioanalyzer results gives us both concentration and molarity values for various peaks. It also gives a total concentration and molarity for the 200-1000bp range, which covers most, but not all of the library. I'm assuming we calculate using the molarity values rather than the concentration values, but I'm not sure if we use the total molarity for the 200-1000bp range they provide, or if we need to include everything else as well. That's my first question.


Even they, trying to get rid of the adapter peak is difficult at best with the titration process. Reading through the forum, it appears as though people prefer using a gel extraction to get rid of the excess adapter. However, we don't have the means to obtain or access something like a Pippen Prep; we'd have to do it by hand using a gel extraction kit. I personally have never done this before. In theory, the adapter dimer is ~130bp, and our absolute minimum size goal for the library would be 230bp (adapters + 100bp fragment), if not longer given that I believe the recommended fragment size for the Illumina platform is something like 200-300bp. In this case, would doing a gel extraction by hand be worth doing instead of trying to optimize? In terms of downstream analyses, would there be any issues or concerns to be aware of?

Thanks

[luc]

Hi Anjama,

I would use the Bioanalyzer molarity values generate for ''regions" with the boundaries set according to your preferences.

Adapter dimers can be removed by magnetic-bead cleanups which would be easier (if the dimers are not excessive).
Will you be doing single-end or paired-end sequencing?

[anjama]

Will you be doing single-end or paired-end sequencing?

For the species I'm working with it'll probably be single-end, but other people might do short paired-end reads where they overlap

I have some ligation products left over from our last titration attempt, and we have all the stuff necessary to do a gel extraction, so I'm going to give that a try tomorrow because it's cheap to do and we have no other use for these ligation products at this point.

I'm only superficially familiar with how magnetic-bead cleanups work, but have no experience with it. We do have access to the hardware necessary, and I think all we need is the beads themselves, so I'll look into that as well.