Clarification on “#matched fragment ions” calculation in TopPIC output

19 views
Skip to first unread message

Ayesha Feroz

unread,
Oct 10, 2025, 4:12:52 AM (4 days ago) Oct 10
to TopPIC

Hello,

I am working on validating TopPIC’s fragment ion matching by comparing its results with my own peak-matching script, and I would like to better understand how the “#matched fragment ions” column in the PRSM output file is computed.

Could you please clarify:

  1. Does TopPIC include only b/y ions, or also a/c/z (and possibly internal ions)?

  2. Are multiple fragment charge states (z = 1, 2, …) included in the count?

  3. Does TopPIC consider losses such as –H₂O or –NH₃ when reporting matched fragment ions?

  4. Does TopPIC always use the user-specified fragment mass tolerance, or is there additional internal recalibration that can affect matching?

  5. Are there any filters (e.g., intensity thresholds, deisotoping, or noise handling) applied before counting a fragment as matched?

I want to make sure my independent matching method is synchronized with TopPIC’s criteria so that I can directly compare my “matched fragment counts” against the values TopPIC reports.

Thank you very much for your help and clarification!

Best regards,
Ayesha

Liu, Xiaowen (Kevin)

unread,
Oct 10, 2025, 10:02:21 AM (4 days ago) Oct 10
to Ayesha Feroz, TopPIC
Hi, Ayesha, 

Thank you for testing TopPIC and sending me your questions. 

  1. Does TopPIC include only b/y ions, or also a/c/z (and possibly internal ions)?

For HCD and CID spectra, TopPIC considers only b/y ions. For ETD spectra, TopPIC considers only c and z-dot ions. 

  1. Are multiple fragment charge states (z = 1, 2, …) included in the count?

Yes. For example, if there are two ions B10 with charge +9 and +10, they are counted as two ions. 

  1. Does TopPIC consider losses such as –H₂O or –NH₃ when reporting matched fragment ions?

No. But you can post processing methods and visualization tools, like TopMSV, to find other types of matched ions.
 
  4. Does TopPIC always use the user-specified fragment mass tolerance, or is there additional internal recalibration that can affect matching?

TopPIC uses the user-specified fragment mass tolerance and two other methods to determine matched fragment ions: (1) Internal recalibration and (2) If the fragment mass > 5000 Da, +/- Da deconvolution errors are allowed. 

  1. Are there any filters (e.g., intensity thresholds, deisotoping, or noise handling) applied before counting a fragment as matched?

Fragment ion filtering was performed in TopFD. For example, you can use EnvCNN score to filter fragment masses. In the default setting, TopFD will estimate the total number of b and y ions based on the precursor mass and use this number to filter fragment mass using EnvCNN or MS-Deconv scores. 

Please let me know if you have other questions. 

Xiaowen (Kevin)

From: top-dow...@googlegroups.com <top-dow...@googlegroups.com> on behalf of Ayesha Feroz <ayeshafe...@gmail.com>
Date: Friday, October 10, 2025 at 3:12 AM
To: TopPIC <top-dow...@googlegroups.com>
Subject: [ms-align-plus-pipeline] Clarification on “#matched fragment ions” calculation in TopPIC output

External Sender. Be aware of links, attachments and requests.

--
You received this message because you are subscribed to the Google Groups "TopPIC" group.
To unsubscribe from this group and stop receiving emails from it, send an email to top-down-topp...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/top-down-toppic/2918e00e-e28a-4ee8-8405-6ad4c2ded272n%40googlegroups.com.
Reply all
Reply to author
Forward
0 new messages