I would say that 1% FDR is strongly recommended and that is what most researchers do. You should always highlight what is a % of FDR in method section. It depends on reviewer (only, not a journal) but personally I would ask in a review process why you have decided to use such high FDR.

Hi,

Are you referring to peptide/protein quantification or analysis of deferentially expressed proteins?

peptide-to-spectrum matches (PSMs) are usually compiled at 1% FDR , regarding data analysis - FDR score can be 0.01 or 0.05.

You can check out the following paper, which covers the basic computational workflows for quantitative label-free proteomics

"The MaxQuant computational platform for mass spectrometry–based shotgun proteomics"

doi:10.1038/nprot.2016.136

Good luck

Elinor

Michal Gorka Dear Dr. Gorka, thank you so much for taking time to read and respond to my questions. So I am using Mascot to identify proteins from a bacterial proteome digest with acquired with a QToF. I realized the column load was pretty low, but at the end of the search I get to see around 700-800 proteins and when I check the FDR its at ~20%. When I adjust it to 1%, the number of proteins comes out to be 60-70. I am using a decoy search as well. I have not however tried any number in between the default FDR and 1%, but my guess it will still be 10 fold less in terms of the number of proteins. I have been pointed out my s proteomics specialist that at 1% FDR, my total number of proteins are going way down as my MS/MS signal is not that great and I am suffering from low sensitivity, which is why I am planning to check the optimum column load and hoping at a optimum load, I will get more proteins at 1% FDR. All these for bottom-up, label free, untargeted proteomics.

Elinor Aviv-Sharon Dear Elinor, thank you so much for taking time to read and respond to my inquiry. I am referring to protein identification using Mascot. I am just trying to optimize different instrumental parameters and the chromatography to increase the number of proteins at a stringent FDR value. Once this goal is achieved I was planning to go for quantitation using clinical samples. But the problem I am facing is that my protein ID is below 100 at 1% FDR, which is why I thought if higher than 1% FDR is acceptable or not. I do understand that, at higher FDR%, confidence on the data goes down as some will be random match and not true match. Thank you so much for referring the article, I will read it to get educated better. Thank you so much.