Glycogen Vs. Ligation

[cometarossa]

Dear all,
I've been fighting for a while with an ineffective library preparation for Illumina sequencing. Just to introduce the problem, in this protocol I have to pass trough steps of digestion, ligation (NEB T4), purification and amplification on freshly extracted plant DNA.
As you probably imagined by now, I'm not obtaining amplification whatsoever. I've proceeded to all sort of troubleshooting, but no results.
Yet, I'm going to try to make my question as specific as possible:

In order to clean the samples (contamination by SIGMA kit), I precipitated my DNA in EtOH using glycogen prior to library preparation: is anybody aware of problems that glycogen can cause with enzimatic reactions? (digestion/ligation) My doubt right now is that, even if stated otherwise, it could interfere somehow.
If yes, does anybody know how to remove it?

Thanks for any help!
Matteo

[C.R.]

I do not think that glycogen is the major problem, since I use it as well during ethanol precipitation after phenol extraction in an illumina library prep. protocol. But I cannot exclude that there might be a harmful effect if you use too much of it.

[pbluescript]

Glycogen can interfere with library prep, and there was a paper a couple years ago that mentioned detectable nucleic acid contamination in commercially available glycogen.

http://www.biotechniques.com/multimedia/archive/00074/BTN_A_000113276_O_74564a.pdf

I use linear acrylamide when I need a co-precipitant, and it has always worked well.

As for removing the glycogen, I would expect people who extract DNA from plants or liver would have some good protocols you can try.

[cometarossa]

Many thanks for your answers!
Actually I'm not that bothered by DNA contaminants, since at the moment I'm still trying to get some DNA from my PCR
Somewhere on the net, however, I read somebody's post stating that glycogen concentration should not be a problem if below 8ng/ul. No idea on where this number comes from. However, the highest concentration i have to do with should be around 1 or 2ng/ul. Still, no references to certain data.
Thanks again for your help, I'll let you know!
Cheers
Matteo