PredMHC: An Effective Predictor of Major Histocompatibility Complex Using Mixed Features

Chen, Dong; Li, Yanjuan

doi:10.3389/fgene.2022.875112

Cited by 4 publications

(2 citation statements)

References 65 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Based on this we hypothesize that future studies with more comprehensive patient cohorts from diverse genetic backgrounds will provide further insights into improving immunopeptidome data acquisition parameters for diverse HLA allele populations. We believe that deep single shot immunopeptidome datasets generated on the SCP will improve HLA binding and presentation prediction algorithms that are presently trained using predominantly ion trap and Orbitrap data (26,(35)(36)(37). Similarly, CCS prediction has been shown to increase confidence in low abundant peptide identifications of complex samples, but so far have been mainly trained on tryptic peptides (20).…”

Section: Discussionmentioning

confidence: 99%

Sensitive, high-throughput HLA-I and HLA-II immunopeptidomics using parallel accumulation-serial fragmentation mass spectrometry

Phulphagar¹,

Ctortecka²,

Vaca³

et al. 2023

Preprint

View full text Add to dashboard Cite

Comprehensive, in-depth identification of the human leukocyte antigen HLA-I and HLA-II tumor immunopeptidome can inform the development of cancer immunotherapies. Mass spectrometry (MS) is powerful technology for direct identification of HLA peptides from patient derived tumor samples or cell lines. However, achieving sufficient coverage to detect rare, clinically relevant antigens requires highly sensitive MS-based acquisition methods and large amounts of sample. While immunopeptidome depth can be increased by off-line fractionation prior to MS, its use is impractical when analyzing limited amounts of primary tissue biopsies. To address this challenge, we developed and applied a high throughput, sensitive, single-shot MS-based immunopeptidomics workflow that leverages trapped ion mobility time-of-flight mass spectrometry on the Bruker timsTOF SCP. We demonstrate >2-fold improved coverage of HLA immunopeptidomes relative to prior methods with up to 15,000 distinct HLA-I and HLA-II peptides from 4e7 cells. Our optimized single-shot MS acquisition method on the timsTOF SCP maintains high coverage, eliminates the need for off-line fractionation and reduces input requirements to as few as 1e6 A375 cells for > 800 distinct HLA-I peptides. This depth is sufficient to identify HLA-I peptides derived from cancer-testis antigen, and novel/unannotated open reading frames. We also apply our optimized single-shot SCP acquisition methods to tumor derived samples, enabling sensitive, high throughput and reproducible immunopeptidome profiling with detection of clinically relevant peptides from less than 4e7 cells or 15 mg wet weight tissue.

show abstract

Section: Discussionmentioning

confidence: 99%

Sensitive, high-throughput HLA-I and HLA-II immunopeptidomics using parallel accumulation-serial fragmentation mass spectrometry

Phulphagar¹,

Ctortecka²,

Vaca³

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The sequential features were generated using the BLOSUM50 matrix. We used iFeature Python library to generate the physicochemical properties [3,4]. A hybrid model was created with a recurrent neural network (RNN) for sequential features whose latent output combined with physicochemical properties was to fed into a deep feedforward network.…”

Section: Methodsmentioning

confidence: 99%

Combining Sequence Similarity with Physicochemical Properties to Predict Binders for MHC-II Molecules

Chen

2023

Caring Is Sharing – Exploiting the Value in Data for Health and Innovation

View full text Add to dashboard Cite

Prediction of binders of the major histocompatibility complex class II (MHC-II) molecules is critical for T cell immunogenicity. As protein-protein interaction also relies on physicochemical properties, we aim to build a novel model combining sequence information and the physicochemical properties of proteins. Our research used data from the NetMHCIIpan 3.2 study. Features include BLOSUM50 and the physicochemical properties from iFeature Python package. We created a hybrid model of recurrent neural layers and feedforward layers. The final Area Under the Receiver Operating Characteristics (AUROC) on the test data was 0.755.

show abstract

Sensitive, High-Throughput HLA-I and HLA-II Immunopeptidomics Using Parallel Accumulation-Serial Fragmentation Mass Spectrometry

Phulphagar¹,

Ctortecka²,

Jacome³

et al. 2023

Molecular & Cellular Proteomics

View full text Add to dashboard Cite

PredMHC: An Effective Predictor of Major Histocompatibility Complex Using Mixed Features

Cited by 4 publications

References 65 publications

Sensitive, high-throughput HLA-I and HLA-II immunopeptidomics using parallel accumulation-serial fragmentation mass spectrometry

Sensitive, high-throughput HLA-I and HLA-II immunopeptidomics using parallel accumulation-serial fragmentation mass spectrometry

Combining Sequence Similarity with Physicochemical Properties to Predict Binders for MHC-II Molecules

Sensitive, High-Throughput HLA-I and HLA-II Immunopeptidomics Using Parallel Accumulation-Serial Fragmentation Mass Spectrometry

Contact Info

Product

Resources

About