I have the same issue; I'm running a single search with Comet and then PeptideProphet, then copying the output pepXml and running XPress separately for H/L and M/L. This gives me answers, but I'd have a lot more confidence in an enhanced XPress that was aware of all three labels (in my case, light-medium-heavy SILAC) and was choosing scan extents appropriately across all the labels.
If anyone's mulling over this kind of enhancement, it would be really valuable! Thanks, Damon On Friday, April 24, 2015 at 3:41:02 AM UTC-7, Alejandro wrote: > > Hi all, > > Well, so far I've done static searches for each label or doing the > variable search as Brian commented, followed by Peptide/ProteinProphet to > combine the files, and it somehow works, of course it get's a little bit > tricky to handle all of the files. I haven't tried so far used Comet in > this setup, I will try and see, looks promising. Regarding the XTandem, I > have indeed used it like with several modifications, but I had forgotten > that setup, I will try it and see if the ID numbers goes down, compared to > doing the three searches. > > As for the quantification this is where it gets more troublesome. The data > that I'm using is Qexactive data, and so far I haven't had so good > experience using ASAP ratio, I think we discussed it in a post, was really > slow in 4.7, but since TPP 4.8 it is faster, but I see better results using > Xpress. So, my strategy so far has been to run it in pairs, do Xpress with > Peptide/Protein Prophet to Heavy/Medium and then Heavy/Light and deduce the > M/L by dividing the two. The problem so far is in this last step. How to > normalize the data (substracting median, etc) when the heavy part is gone. > Plus how to handle when one ID is not found in one channel or when the > quantification also didn't work in one/two channels. > > Any ideas? > > Alejandro > > > > On Tuesday, April 21, 2015 at 11:55:58 AM UTC+2, Alejandro wrote: >> >> Hello all, >> >> I have some experiments in which the samples were triple dimethyl >> labeled. Does any of you have any tips on how to analyze these data with >> later version of TPP? >> >> I looked through the list but only could find a couple of posts. >> >> So far my ideas would be to do: >> >> 1. Static searches of Heavy, Medium and Light and then analyze H/M, and >> H/L, and deduce the M/L by dividing the ratios of (H/L)/(H/L) ? >> >> So far I'm doing that, however after Peptide/ProteinProphet the results >> of each vary quite abit and might end up with peptides/proteins with a >> ratio in just two conditions and not in the three. Usually this also >> happens in MaxQuant in triple labeled but only in very few hits. >> >> 2. Would it be possible to run a variable search for all three >> modifications in XTandem in one run, Heavy, Medium, Light and then just >> process H/M and H/L with Peptide/ProteinProphet? >> >> >> Maybe some of you had any experience with this, any tips or help would be >> appreciated! >> >> Thanks, >> >> Alejandro >> > -- You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at http://groups.google.com/group/spctools-discuss. For more options, visit https://groups.google.com/d/optout.
