Could you use 18% more peptide IDs with your normal confidence level (assuming a 1% FDR)?
Wait. I can one up this. Could you use 23% more ID's at a 0.1% FDR?
Is it finally time for us to seriously look at spectral libraries again? I'm gonna say it is. I also think this paper is another good argument to support it.
If you've also been doing this a while you might think "hey, we tried this, but the libraries weren't good enough"
Okay -- so there are some people, I think they're mostly in Germany (but we won't hold that against them) and they are synthesizing something like every human peptide. If you haven't seen it, I recommend you check it out here.
NIST also hasn't been sitting around doing nothing all these years either. They've already made a huge libraries of increasing quality -- and they've already made spectral libraries of the first ProteomeTools release.
You can check out what NIST has available these days here.
Does anyone know if MSPepSearch works in the IMP-Proteome Discoverer? I'm trying it out, but I won't know for sure until the demo key expires in 54 days....
No comments:
Post a Comment