Custom modification search

[treko90]

Hi, is it possible to somehow pass the customized Unimod database to PepQuery for modification search? For example, Unimod file contains various isotopic labels which are not relevant for my experimental setting and sometimes identification is confused by them.

I'm asking because I see that the PepQuery version on the github refers to unimod.xml file in the resources directory but there is no such directory or file in in the latest downloadable PepQuery

[wenbostar]

It is packaged inside the jar file. I could add a command line option to use a user-defined unimod file.

[treko90]

Thank you.

Another question - I want to search for a potential SNP in my proteomic data.
My reference sequence is TGHDFPIDTK and its variant should be TGHEFPIDTK.
I want to distinguish between TGH(D->E)FPIDTK and TGHDFPI(D->E)TK so I've searched for TGHEFPIDTK and put -aa tag to the pepquery run command. However, this causes that peptide is reported as not confident (because its aa substitution of the reference peptide) but TGH(D->E)FPIDTK suggestion (passed other tests) can be found in ptm_detail table without any other candidate. Can such not-confident peptide be actually considered as confident?

[wenbostar]

For the spectrum matched to TGHEFPIDTK, you could compare it with predicted spectra from TGHEFPIDTK and TGHDFPI(D->E)TK using PDV to see which one gives you higher correlation.

[treko90]

But wouldnt PepQuerys UMS with -aa option automatically provide the scores for both variants if they were competitive?

[wenbostar]

If "-fast" is NOT set, you will be able to find the scores for compared reference peptides for a given target peptide in file ptm_detail.txt.