Hi Thomas, I am not sure what is happening but if you could provide the dataset files for the two search results I can investigate. PeptideProphet does calculate different models for the different crosslink types.
Thanks, -David On Thu, Sep 12, 2019 at 8:59 AM Thomas Gossenreiter < [email protected]> wrote: > Hi, > > I have used Kojak 1.6.1 in the TPP pipeline and got some strange results > (very similar ms2 spectra from one peptide with similar Kojak scores got > very different probabilities when running peptide prophet). I have plotted > the Kojak score vs. the peptide prophet probability for target and decoy > hits and there is hardly any correlation between the kojak score and the > probability. > > I have repeated the search Kojak version that comes along with TPP (1.5.5) > using the same settings except for the top_count parameter. I think this > looks how it should, except for the cross-links which are scored very well > by Kojak but get a bad peptide prophet probability. Is this the case when > one peptide has a very good Kojak sub score and the other one a very bad > score? > > Does version 1.6.1 work for you or do you have any idea what is happening > here? I attached a ppt file with the figures. > > I also got a general questions: Is peptide prophet calculating FDRs > separately for cross-linked, loop-linked and normal peptides? > > Kind regards, > Thomas > > > > > -- > 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 view this discussion on the web visit > https://groups.google.com/d/msgid/spctools-discuss/12f12195-f115-41e5-a196-b53e18e170df%40googlegroups.com > <https://groups.google.com/d/msgid/spctools-discuss/12f12195-f115-41e5-a196-b53e18e170df%40googlegroups.com?utm_medium=email&utm_source=footer> > . > -- 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 view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/CAGJJY%3D-1%2B%3Da%3DZ9qXHcgPc3w68gto_9uWCviJsWwEqX08%2BVVi_A%40mail.gmail.com.
