Tide-search Xcorr values seem wildly different from Thermo "PSMS" software values

[Michael Pique]

I am running Tide-search on ms2 files which our collaborator has run Thermo's PSMS software on.  The Thermo Xcorr values are like we have seen for many years, between .5 and 2 or 3.  But the Tide values are much smaller (0.6 max) and about half of them are negative.  A scatter plot show no correlation. I feel I am missing some setting but have tried  some without success.  I will attach the parameter file and a screenshot of the scattergram and the beginning of the tables.  Here is the log:

NFO: CPU: Luna.local

INFO: Crux version: 4.1-4ff75ee-2022-09-28

INFO: Wed Jan 18 12:41:40 PST 2023

COMMAND LINE: crux tide-search --use-tailor-calibration T --output-dir tide9-Elite202212a74-OPSD Elite202212a74.ms2 OPSD

INFO: Beginning tide-search.

INFO: Running tide-search...

INFO: Number of Threads: 1

INFO: Reading index OPSD

INFO: Read 1 target proteins

INFO: Reading auxiliary locations.

INFO: Reading peptide_header.

INFO: Converting Elite202212a74.ms2 to spectrumrecords format

INFO: Elapsed time starting conversion: 0.00688 s

INFO: Converting ms_level 2 ... 

INFO: Reading spectrum file tide9-Elite202212a74-OPSD/Elite202212a74.ms2.spectrumrecords.tmp.

INFO: Read 5505 spectra.

INFO: Starting search.

INFO: 1000 spectrum-charge combinations searched, 18% complete

INFO: 2000 spectrum-charge combinations searched, 36% complete

INFO: 3000 spectrum-charge combinations searched, 54% complete

INFO: 4000 spectrum-charge combinations searched, 73% complete

INFO: 5000 spectrum-charge combinations searched, 91% complete

INFO: [Thread 0]: Deleted 0 precursor, 0 isotope and 0 out-of-range peaks.

INFO: [Thread 0]: Retained 100% of peaks.

INFO: Time per spectrum-charge combination: 0.000476 s.

INFO: Average number of candidates per spectrum-charge combination: 1.388193 

INFO: Elapsed time: 2.62 s

INFO: Finished crux tide-search.

INFO: Return Code:0

[William S Noble]

This is indeed strange since we expect to see some correlation. However, I noticed that you are not actually plotting the Tide XCorr -- you are plotting the Tide Tailor score. Can you set --use-tailor-calibration F and re-do the plot?

Note that we're shifting over to doing most of our support via the github issue tracker, so I've copied this over there:

https://github.com/crux-toolkit/crux-toolkit/issues/624

On Thu, Jan 19, 2023 at 10:31 AM Michael Pique <terrat...@gmail.com> wrote:

I am running Tide-search on ms2 files which our collaborator has run Thermo's PSMS software on.  The Thermo Xcorr values are like we have seen for many years, between .5 and 2 or 3.  But the Tide values are much smaller (0. 6 max) and about half

ZjQcmQRYFpfptBannerStart

This Message Is From an Untrusted Sender

You have not previously corresponded with this sender.

See https://itconnect.uw.edu/guides-by-topic/email-calendaring/protecting-your-email/email-tags/ for additional information.

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 on the web visit https://groups.google.com/d/msgid/crux-users/05887b1f-82ea-4a76-9c06-0a6119eb8872n%40googlegroups.com.

[Michael Pique]

Bill, thanks for looking at this so fast.  I sent Andy Lin a description of what I did to extract the sequences and the Xcorrs from the PSMS and Tide output files.

I often make mistakes - but looking again at the "tide-search.target.txt", I'm pulling out and plotting the ninth tab-separated field, "xcorr score" not the tenth, "tailor score". I just reran with "--use-tailor-calibration F" and get precisely the same Xcorr values, only difference I see is that the "tailor score" header and data field are gone.  I can pull out the first hundred or so lines of each and send if helpful.