Pre-trained models
jalew188
released this
28 Jun 13:30
·
691 commits
to main
since this release
Pre-trained MS2/RT/CCS models
Change log:
[2022-11-28] For timsTOF data, fragment intensities for training are no longer extracted from MaxQuant's msms.txt. We extracted fragments of AlphaPept PSMs against the ms_data.hdf, this should be more stable than msms.txt for timsTOF. The previous model is still available on https://datashare.biochem.mpg.de/s/RDSZlAnCRvxfLfk.
[2024-06-29] Add the pretrained LSTM-based HLA-I model (hla_model.zip) for peptide binding prediciton.