Yes, "Refine > Reintegrate > Add" is how you tell Skyline to train a peak scoring model and eventually calculate FDRs.

You should take a look at the Advanced Peak Picking tutorial:
https://skyline.ms/wiki/home/software/Skyline/page.view?name=tutorial_peak_picking

That tutorial will show you how to add decoys to your document, extract chromatograms and then train a peak scoring model.
-- Nick

Hi Nick,

Thank you for your response. My understanding is that after "Reintegrate", I will get peptide q value which represents peptide FDR rate in my export file, right?

Another question I have is that how do you prepare the peptide/transition list for peptide identification with FDR control? If I am only interested in a small group of proteins in my samples, should I still import the whole proteome fasta file to generate the peptide/transition list so that it provides an accurate estimation of FDR? Or can I just import a fasta file of proteins that I am interested in, and the FDR will still be correct? The reason I am asking is because if I import the whole proteome, the Skyline file will be huge and the processing will be extremely slow. Do you have any solution to it?

Thank you,
Zhenze

You can use a subset FASTA. You should also consult the DIA/SWATH tutorials on this page:

https://skyline.ms/tutorial_dia_swath.url

If you go too small that could impact your statistics, but the original mProphet paper suggests that this works with relatively small target lists:

https://pubmed.ncbi.nlm.nih.gov/21423193/