Zeng, Wen-Feng

I have checked the Raw/MS1_scans table and Raw/MS2_scans tables on both win and mac, the values are identical. So the pyrawfilereader is not the problem. But the file size are...

In fasta.py, ``` def get_frag_dict(parsed_pep, mass_dict): frag_dict = {} frag_masses, frag_type = get_fragmass(parsed_pep, constants.mass_dict) ... ``` the mass_dict argument is never used as well.

We are currently working on this, check https://github.com/MannLabs/alphaviz/blob/peptdeep_latest/nbs/tutorial_DIANN_alphabase.ipynb

And plot_elution_profile() calculates b/y ions inside this function with mass_dict, making it hard to plot elution for modifications not in mass_dict, I suggest to calculate b/y ions into peptide_info outside...

Hi @michaelsteidel86 AlphaPeptDeep does not support burker/sciex data for ms2 training as we don't have a good centroiding algorithms for TOF-MS. We plan to support this, hope it will not...

You can try to use frag-pipe's output spectral library for MS2/RT/CCS training. In APD GUI, you can select speclib_tsv for both PSM and RAW

It turns out that this is not “catastrophic forgetting”, this is due to low matching quality of DIA data.

Could you please check diann report file? "Modified.Sequence" must be in the header

Thank you for your report, i will investigate this soon.

No, we did not train model for H2O and NH3 losses