Different results with different file order with mass accuracies and scan windows set

[jjjotto]

I see this question and answer from 2021 and wanted to check if setting the scan window and mass accuracies should result in the same results independent of file order. https://github.com/vdemichev/DiaNN/issues/79

I'm pasting the DIA-NN commands used for processing below. If something is incorrect, please let me know. We are new to the software.

If the scan window and mass accuracies are set in both commands, should the results be different if the order of the files is different? The results are not drastically different.

Thank you in advance!

C:\DIA-NN\2.3.1\diann.exe --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307753_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307752_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307751_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307750_postspe.raw --lib Y:\temporary_files\FASTA\DIA-NN_SpecLibs\MOUSE_reviewed_2025_12_31.predicted.speclib --threads 16 --verbose 1 --out D:\DIANN_runs\1307750-53_post_diforder\1307750-53_post_diforder.parquet --qvalue 0.01 --out-lib C:\DIA-NN\2.3.1\report-lib.parquet --gen-spec-lib --reannotate --fasta Y:\temporary_files\FASTA\MOUSE_reviewed_2025_12_31.fasta --matrices --met-excision --min-pep-len 7 --max-pep-len 30 --min-pr-mz 380 --max-pr-mz 980 --min-pr-charge 1 --max-pr-charge 6 --min-fr-mz 150 --max-fr-mz 1500 --cut K*,R* --missed-cleavages 2 --unimod4 --window 5 --mass-acc 10 --mass-acc-ms1 4 --pg-level 1 --no-prot-inf --rt-profiling --reanalyse

C:\DIA-NN\2.3.1\diann.exe --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307750_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307751_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307753_postspe.raw --f Y:\msdata\2026\2026_02\OAZ1\OAZ1_1307752_postspe.raw --lib Y:\temporary_files\FASTA\DIA-NN_SpecLibs\MOUSE_reviewed_2025_12_31.predicted.speclib --threads 16 --verbose 1 --out D:\DIANN_runs\1307750-53_post_NEW\1307750-53_post_NEW.parquet --qvalue 0.01 --out-lib C:\DIA-NN\2.3.1\report-lib.parquet --gen-spec-lib --reannotate --fasta Y:\temporary_files\FASTA\MOUSE_reviewed_2025_12_31.fasta --matrices --met-excision --min-pep-len 7 --max-pep-len 30 --min-pr-mz 380 --max-pr-mz 980 --min-pr-charge 1 --max-pr-charge 6 --min-fr-mz 150 --max-fr-mz 1500 --cut K*,R* --missed-cleavages 2 --unimod4 --window 5 --mass-acc 10 --mass-acc-ms1 4 --pg-level 1 --no-prot-inf --rt-profiling --reanalyse

[vdemichev]

Hi,

Identification numbers without MBR will be the same, if you don't filter by any global confidence metrics. Quantites or any global confidence metrics may be different. With MBR results are likely to be quite different. In the above you are using MBR.

Best,
Vadim