Max input file size limit for comet search?

Skip to first unread message

Steven He

Sep 17, 2021, 1:58:10 AM9/17/21
to spctools-discuss

I'm a fairly new user to TPP and this sort of analysis in general and had a question about input file sizes during Comet Search.
I have downloaded some MS data from a publication via an online database  in .mzXML format (the .raw files weren't provided) and have been trying to run a peptide database search using the Comet software with the default param settings. However I consistently get the message "Warning - no spectra searched" and no output file is created (screenshot attached). The mzXML file itself is actually quite large (~26GB) and I was wondering if the size of the input file itself might be the issue? In the case where the input files would be quite large, is there a potential workaround?

Apololgies if I've left out any critical information here

Thanks in advance,

TPP error.png

Jason Winget

Sep 17, 2021, 1:01:59 PM9/17/21
to spctools-discuss
Hi Steven, what do your comet parameters look like?
I'm not sure if this is the problem, but for files with many scans it is good to set the `spectrum_batch_size` parameter to something reasonable, like 10000. The instructions are found in the "fragment ions" comments section of the default parameters file.

Eric Deutsch

Sep 17, 2021, 1:11:18 PM9/17/21
to, Eric Deutsch

I’m thinking that a very large mzXML file with “UDMSE” in it might be DIA data, and thus Comet is not the right tool? Could this be MS^e data instead of DDA data?

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
To view this discussion on the web visit

Steven He

Sep 19, 2021, 8:57:21 PM9/19/21
to spctools-discuss
Thanks for the responses,

In the comet parameter file, the 'fragment_bin_tol' value is set to 1.0005 and 'spectrum_batch_size' is set to 15000. I tried running the data using 'spectrum_batch_size' set to 10000 but still had the same issue, which I think Eric has addressed. I will keep in mind to check this setting in future though.

Thanks Eric, I think that was the issue; the data was acquired using a DIA setup and not DDA. I haven't done much work with DIA data before so I will do some reading online and have another crack at it.

Thanks again for your responses!
Reply all
Reply to author
0 new messages