problems with sample degradation in the lab

[mboth]

I have experienced some problems in the lab recently, and I am interested if anybody has any comment or suggestions.
We have noticed lower than usual yields for library preps. Now some of the data has come back, and there is a high level of duplication in the samples. I would normally expect this from a sample which has been over-amplified.
The problem first occurred in RNA-seq libraries, but now also does in any other kind of library prep we do.
I am thinking that something in my lab is degrading RNA and DNA alike as I am preparing the libraries.
My number one suspect is the DNA Zap solutions that I frequently use in order to avoid cross-contamination of projects.
Has anybody ever seen this happen? How potent is the DNA Zap? What are other people using?
I appreciate any comment you might have.

[ksawatzki]

One basic thing would be to make sure all of your reagents, particularly before amplification, are RNase and DNase free, and you work in a hood treated with RNase/DNase inhibitor. If you've been using the same bottle of water/diluent, I'd put that aside and use it for non-sensitive things and use a brand new bottle.

Also, keep in mind that degradation might be upstream of your work. Were the samples collected, processed and stored together with the same number of freeze-thaws? These can all contribute to degradation.

[pmiguel]

Sounds like you are drawing a conclusion (that your samples are degraded) without any direct evidence that this is the case. Are your samples degraded? How about checking that directly?
That said, I've seen people go overboard with that "zap" stuff -- to the point that it was clogging up pipettors, etc. If the residue is visible on your equipment, some of that zap dust could be getting into your samples and that would probably make it difficult to construct libraries from them.

--
Phillip

[mboth]

Thank you for your thoughts so far.
Of course I had a great clean up, and hope that this will resolve the issue.
The data that comes back shows that some samples were victim of over-amplification during the PCR (as it might happen when only a few molecules survive the "zapping"). My libraries also show a very low and partially uneven coverage of the mapping. There seems to be a particular problem with GC-rich regions as well, not much is mapping to those.
Any ideas what could cause this?