Spliting FAIMS DIA files and combining proteins IDs of different CV

[sagigord]

Hi, I'm new to DIA-NN and interested in analysis of DIA-FAIMS files. Basically each sample has been run with 4 different CVs , So we have 4 files per each sample. I would be happy to ask 2 different things regarding DIA with FAIMS.

1. Does anyone find a better option for splitting files for CVs which is not manually based on Thermo freestyle?
2. Can I run maxLFQ on the report and get normalized pg_matrix and then just average by median the protein intensities of 4 files (originated from 1 sample) as a group?

Thanks a lot! Best Sagi

[vdemichev]

Hi Sagi,

1. Without splitting quantification is unlikely to work properly. Identification - not sure.
2. MaxLFQ should be run only on precursor IDs measured with the same CV value. Otherwise the results would not make sense. So need to run it separately for different CVs.

Best, Vadim

[sagigord]

Hi Vadim, Thank for your reply. I asked about other tool then splitting manually , because until now I splitted manually and I'm having 400 files... Regarding LFQ normalization, I used 4 cvs per each sample. Do you mean I need to normalize all the files of CVx first, second normalize all files of CVy etc. I would be happy if you can share some R code or something I can get help from. In addition , after normalized values of each protein per CV, how should I combined the 4 values to 1 column per sample? seems to me that just average protein intensity by mean is not the correct way , no?

I would really appreciate your help. We got very nice results and will be happy to continue for the analysis steps. Thanks a lot! Sagi

[KlemensFroehlich]

hi sagi, Can you simply perform 4 completely separate analyses and look at the correlation of the results? Correlation of fold changes of proteins would be very interesting to see. Best Klemens