compomics / ms2rescore

Modular and user-friendly platform for AI-assisted rescoring of peptide identifications
https://ms2rescore.readthedocs.io
Apache License 2.0
39 stars 14 forks source link

Correctly handle peptides with invalid residues (BJOUXZ) #106

Closed RalfG closed 2 months ago

vrkosk commented 3 months ago

A concrete use case this happens is with Mascot search results. For example, this mzIdentML snippet is from exporting results of a search against SwissProt:

    <Peptide id="peptide_28284_10">
      <PeptideSequence>APLLVPGFCVAXNQPTMR</PeptideSequence>
      <Modification location="0" monoisotopicMassDelta="144.102063">
        <cvParam accession="UNIMOD:214" name="iTRAQ4plex" cvRef="UNIMOD" />
      </Modification>
      <Modification location="9" residues="C" monoisotopicMassDelta="45.987721">
        <cvParam accession="UNIMOD:39" name="Methylthio" cvRef="UNIMOD" />
      </Modification>
      <SubstitutionModification location="12" originalResidue="X" replacementResidue="A" />
    </Peptide>
RalfG commented 2 months ago

Fixed in https://github.com/compomics/ms2rescore/commit/8613a38a8ea88b3c51591d9437aa432ef9d8dd39.