software threw 2 samples from quant

[Dmorgen]

I'm running a 90 sample analysis, but the output contains only 88 columns. does the software throw outliers?

D.

[rmillikin]

it doesn't remove outlier files; maybe there were no PSMs observed below 1% FDR for those two runs? can you check on that? If this is the case I can report an error message if this happens (but still continue with the analysis, if desired)

[Dmorgen]

There is definitely data there and the spectrum is quite rich. The only thing, I’m using the MBR function here – no ID’s specifically from these files. D. From: Robert Millikin <[email protected]> Sent: Tuesday, March 31, 2020 8:35 PM To: smith-chem-wisc/FlashLFQ <[email protected]> Cc: David Morgenstern <[email protected]>; Author <[email protected]> Subject: Re: [smith-chem-wisc/FlashLFQ] software threw 2 samples from quant (#89) it doesn't remove outlier files; maybe there were no PSMs observed below 1% FDR for those two runs? can you check on that? If this is the case I can report an error message if this happens (but still continue with the analysis, if desired) — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub<#89 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AIQ5VINROIM2JNQF6R6D5LDRKISUBANCNFSM4LXZDTWA>.

[rmillikin]

MBR currently uses identifications in common between two files to create the relative retention time gradient. if a file has no IDs, nothing in the file can be quantified currently. that will probably change at some point, through "blind" feature finding so you can do this kind of thing. it's not currently supported, though