Re: [spctools-discuss] Crosslink FDR from PeptideProphet

[David Shteynberg]

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 for each type of Kojak result.

Thanks for your question!

-David

On Wed, Sep 16, 2020 at 12:22 PM Lindsey Ulmer <ulmerli...@gmail.com> wrote:

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 correspond to crosslinked results. I've attached a link to the corresponding pepxml below.

https://drive.google.com/file/d/1hadnfjsFVCTMVqirdcC2jCeEuUvzVjAt/view?usp=sharing

Thanks!

Lindsey

--
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 spctools-discu...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/1352899f-62a4-43fa-8ff1-0d6809251e9dn%40googlegroups.com.

[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?

To view this discussion on the web visit https://groups.google.com/d/msgid/spctools-discuss/CAGJJY%3D8fCBMQWB1tSp2vv1D%2Bc1uUxWYn2aQMsLpScy-jBaiS2g%40mail.gmail.com.