generating library from DDA search

[animesh]

Wondering what is the recommended way to generate library from DDA for DIA-NN search? For example i have MaxQuant msms.txt file but not sure how to present it as speclib for DIA-NN search for corresponding DIA data?

Also, if the library is generated from DDA, will the search from DIA be restricted to results from DDA library or DIA-NN can extrapolate new findings?

[vdemichev]

Hi Ani,

FragPipe is awesome :)

how to present it as speclib for DIA-NN search for corresponding DIA data?

There's 'experimental' (means not properly tested) support for msms.txt as input spectral library for DIA-NN. You can convert it to DIA-NN's own .parquet format (specify as input msms.txt and select 'Generate spectral library', don't specify any raw files) and examine it in R ('arrow' package) to see if it looks like it makes sense.

will the search from DIA be restricted to results from DDA library or DIA-NN can extrapolate new findings?

This would be the recommended way: https://github.com/vdemichev/DiaNN/issues/1210#issuecomment-2410984411

Best, Vadim

[animesh]

Thanks @vdemichev , tried but failed 🤪

diann.exe --f "F:\phosDIA\msms.txt"  --threads 12 --out-lib "C:\Users\animeshs\DIANN1p9p1\report-lib.parquet" --gen-spec-lib
DIA-NN 1.9.1 (Data-Independent Acquisition by Neural Networks)
Compiled on Jul 15 2024 15:40:36
Current date and time: Wed Oct 16 11:18:14 2024
CPU: GenuineIntel Intel(R) Xeon(R) CPU E5-2643 v3 @ 3.40GHz
SIMD instructions: AVX AVX2 FMA SSE4.1 SSE4.2
Logical CPU cores: 24
Thread number set to 12
A spectral library will be generated
WARNING: skipping F:\phosDIA\msms.txt - invalid raw MS data format
WARNING: protein inference is enabled but no FASTA provided - is this intended?

0 files will be processed

probably missing some switch(es)?