Global proteome search

[DiegoFGR15]

Dear MSstats team,

I am a fairly new user, and I have a question about the global proteome search.
Do I need to include the PTMs that I am searching as variable modifications for the global proteome search?

I am using the output of FragPipe and DIA-NN using the export-quant command.

And in this case, should I use both .parquet files (enriched and global proteomes) as MsStats inputs?

Thank you very much.
Diego

[tonywu1999]

Do I need to include the PTMs that I am searching as variable modifications for the global proteome search?

No, you don’t need to include PTMs as variable modifications in the global proteome search. Including many variable modifications increases the search space, which can lead to fewer identifications due to multiple hypothesis testing. For global proteome runs, we are only interested in overall protein abundance, so it doesn’t matter whether the peptides are modified or unmodified when representing a protein.

And in this case, should I use both .parquet files (enriched and global proteomes) as MsStats inputs?

Yes, use both .parquet files.

Also, you're using DIANN 2.0+ right? I just noticed MSstatsPTM does not support DIANN 2.0 at the moment (only 1.8), but I can make a quick update to enable support for DIANN 2.0.

[DiegoFGR15]

Thanks for the answers.

If there is the possibility for an update, that would be very useful.

I plan to use DIA-NN 2.3.1. I will use the export-quant command to get the fragment intensities in the .parquet.
Then I will use Arrow and the converter to use MSstats PTM.

Let me know about the update.
BR
Diego Garcia

[tonywu1999]

@DiegoFGR15

That's great to know. We'll get started this week on making the changes and I'll update this thread when it's done.

In the meantime, do you have any sample parquet files we can use as an example to test the functionality / add as an example in the documentation for MSstatsPTM?

Thanks
Tony