Format

Send to

Choose Destination
Nat Biotechnol. 2020 Feb;38(2):199-209. doi: 10.1038/s41587-019-0322-9. Epub 2019 Dec 16.

A large peptidome dataset improves HLA class I epitope prediction across most of the human population.

Author information

1
Department of Biomedical Informatics, Harvard Medical School, Boston, MA, USA.
2
Broad Institute of MIT and Harvard, Cambridge, MA, USA.
3
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA.
4
Harvard Medical School, Boston, MA, USA.
5
Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA.
6
Center for Patient Derived Models, Dana-Farber Cancer Institute, Boston, MA, USA.
7
Division of Neuropathology, Brigham and Women's Hospital, Boston, MA, USA.
8
Department of Cell Biology, Harvard Medical School, Boston, MA, USA.
9
Immunitrack, Copenhagen, Denmark.
10
Department of Pathology, Brigham and Women's Hospital, Boston, MA, USA.
11
Department of Computer Science, Metropolitan College, Boston University, Boston, MA, USA.
12
Broad Institute of MIT and Harvard, Cambridge, MA, USA. nhacohen@mgh.harvard.edu.
13
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. nhacohen@mgh.harvard.edu.
14
Center for Cancer Immunology, Massachusetts General Hospital, Boston, MA, USA. nhacohen@mgh.harvard.edu.
15
Broad Institute of MIT and Harvard, Cambridge, MA, USA. scarr@broadinstitute.org.
16
Broad Institute of MIT and Harvard, Cambridge, MA, USA. cwu@partners.org.
17
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. cwu@partners.org.
18
Harvard Medical School, Boston, MA, USA. cwu@partners.org.
19
Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA. cwu@partners.org.
20
Broad Institute of MIT and Harvard, Cambridge, MA, USA. derin_keskin@dfci.harvard.edu.
21
Department of Medical Oncology, Dana-Farber Cancer Institute, Boston, MA, USA. derin_keskin@dfci.harvard.edu.
22
Harvard Medical School, Boston, MA, USA. derin_keskin@dfci.harvard.edu.
23
Department of Medicine, Brigham and Women's Hospital, Boston, MA, USA. derin_keskin@dfci.harvard.edu.
24
Department of Computer Science, Metropolitan College, Boston University, Boston, MA, USA. derin_keskin@dfci.harvard.edu.

Abstract

Prediction of HLA epitopes is important for the development of cancer immunotherapies and vaccines. However, current prediction algorithms have limited predictive power, in part because they were not trained on high-quality epitope datasets covering a broad range of HLA alleles. To enable prediction of endogenous HLA class I-associated peptides across a large fraction of the human population, we used mass spectrometry to profile >185,000 peptides eluted from 95 HLA-A, -B, -C and -G mono-allelic cell lines. We identified canonical peptide motifs per HLA allele, unique and shared binding submotifs across alleles and distinct motifs associated with different peptide lengths. By integrating these data with transcript abundance and peptide processing, we developed HLAthena, providing allele-and-length-specific and pan-allele-pan-length prediction models for endogenous peptide presentation. These models predicted endogenous HLA class I-associated ligands with 1.5-fold improvement in positive predictive value compared with existing tools and correctly identified >75% of HLA-bound peptides that were observed experimentally in 11 patient-derived tumor cell lines.

PMID:
31844290
PMCID:
PMC7008090
[Available on 2020-06-16]
DOI:
10.1038/s41587-019-0322-9

Supplemental Content

Full text links

Icon for Nature Publishing Group
Loading ...
Support Center