Trouble replicating assign-confidence results

[Erik Johnson]

Hello again,

I'm trying to replicate assign-confidence's q-values (mainly so I can make sure I understand how q-value is calculated). I am using the default "tdc" method. I'm using these two papers for reference: 

• "Assigning Significance to Peptides Identified by Tandem Mass Spectrometry Using Decoy Databases" (2007) Kall, Noble, et al
• "Statistical significance for genomewide studies" (2003) Storey & Tibshirani 
  

The q-value results I'm getting do not match crux assign-confidence's q-values. I'm implementing the algorithm in the appendix of the Storey & Tibshirani paper (screenshot below). I run separate target and decoy searches. I ignore those spectra for which the decoy score beats the target score. Then I set the p-value for PSM i with target score t to be: p_i = #(decoys with score > t) / #decoys. So, the PSMs with target scores greater than any decoy score have p_i = 0 which (by the q-value equations in steps 5 & 6 of the Storey & Tibshirani algorithm) would make q_i = 0 as well. But none of the crux computed q-values are 0. So I must be missing something.

Any help is greatly appreciated. Thanks all!

- Erik

[William Stafford Noble]

Hi Erik,

The method used in assign-confidence is different from the ones in the papers you are looking at.  You might want to look at the Crema paper:

https://analyticalsciencejournals.onlinelibrary.wiley.com/doi/10.1002/pmic.202300084

Cheers.

Bill

On Tue, Oct 21, 2025 at 3:34 PM Erik Johnson <airik.k...@gmail.com> wrote:

Hello again, I'm trying to replicate assign-confidence's q-values (mainly so I can make sure I understand how q-value is calculated). I am using the default "tdc" method. I'm using these two papers for reference: "Assigning Significance to Peptides

ZjQcmQRYFpfptBannerStart

This Message Is From an Untrusted Sender

You have not previously corresponded with this sender.

See https://itconnect.uw.edu/email-tags for additional information. Please contact the UW-IT Service Center, he...@uw.edu 206.221.5000, for assistance.

 

ZjQcmQRYFpfptBannerEnd

--
You received this message because you are subscribed to the Google Groups "crux-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to crux-users+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/crux-users/445db25c-fdd5-4c92-8df1-167971b03a14n%40googlegroups.com.