Re: [spctools-discuss] Crosslink FDR from PeptideProphet

2020-09-16 Thread Lindsey Ulmer
Thanks! I do have less error tables than I would expect. When I ran PeptideProphet, I received the warning: "Mixture model quality test failed for charge (1+)" for multiple charge states. Would that warning account for the missing error tables? On Wed, Sep 16, 2020 at 1:34 PM 'David Shteynberg'

Re: [spctools-discuss] Crosslink FDR from PeptideProphet

2020-09-16 Thread 'David Shteynberg' via spctools-discuss
Hi Lindsey, PeptideProphet will generate a separate FDR prediction table for each type of Kojak result it analysed, and all three should be in the pepXML file. The three types of Kojak hits are unlinked "n/a", self-linked "loop" and cross-linked "xl".PeptideProphet generates a separate model

[spctools-discuss] Crosslink FDR from PeptideProphet

2020-09-16 Thread Lindsey Ulmer
Hi, I'm trying to determine the FDR of crosslinked products from a Kojak and PeptideProphet analysis. There is a second plot in the Predicted Sensitivity and Error Rate chart, which I'm assuming is for the crosslinked products, but I'm having trouble determining which of the error tables