HLAProphet: Personalized allele-level quantification of the HLA proteins
Mumphrey, M. B.; Li, G. X.; Hosseini, N.; Nesvizhskii, A.; Cieslik, M.
Show abstract
Loss of HLA expression in tumor cells is a commonly observed phenotype that is known to be associated with T-cell evasion. Proteogenomic characterizations of the molecular mechanisms underpinning this loss of HLA expression are hindered by the polymorphic nature of the HLA proteins, with most individuals having germline HLA sequences that are highly divergent from the sequences found in standard reference databases. To address this issue, we have developed HLAProphet, an algorithm that utilizes HLA types from paired DNA sequencing data to provide personalized allele-level quantification of the HLA proteins from TMT mass spectrometry data. We show that HLAProphet triples the number of tryptic peptide identifications made by standard reference based approaches, and produces protein expression values that have high concordance with RNA expression and known loss of heterozygosity events.
Matching journals
The top 5 journals account for 50% of the predicted probability mass.
Similar papers in this journal
- Proteoform Identification by Combining RNA-Seq and Top-down Mass Spectrometry 96%
- The Personalized Proteome: Comparing Proteogenomics and Open Variant Search Approaches for Single Amino Acid Variant Detection 95%
- Protein sequencing with single amino acid resolution discerns peptides that discriminate tropomyosin proteoforms 95%
Similar papers in this journal
- PEPerMINT: Peptide Abundance Imputation in Mass Spectrometry-based Proteomics using Graph Neural Networks 94%
- MS2AI: Automated repurposing of public peptide LC-MS data for machine learning applications 94%
- scFeatures: Multi-view representations of single-cell and spatial data for disease outcome prediction 94%
Similar papers in this journal
- Imputation of label-free quantitative mass spectrometry-based proteomics data using self-supervised deep learning 96%
- An adaptive, continuous-learning framework for clinical decision-making from proteome-wide biofluid data 95%
- UbiFast, a rapid and deep-scale ubiquitylation profiling approach for biology and translational research 94%
"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.