Semi-supervised learning for somatic variant calling and peptide identification in personalized cancer immunotherapy

• Published in: BMC bioinformatics Vol. 21; no. Suppl 18; p. 498
• Main Authors: Sherafat, Elham, Force, Jordan, Măndoiu, Ion I
• Format: Journal Article
• Language: English
• Published: England BioMed Central 30-12-2020; BMC
• Subjects: Accuracy; Bioinformatics; Cancer; Cancer immunotherapy; Cancer vaccines; Classification; Customization; Datasets; Deoxyribonucleic acid; DNA; Epitopes - immunology; Epitopes - metabolism; Exome sequencing; Genomes; Humans; Immune system; Immunotherapy; Learning; Machine learning; Model accuracy; Mutation; Neoplasms - therapy; Ovarian cancer; Peptide identification; Peptides; Peptides - analysis; Polymorphism, Single Nucleotide; Positive-unlabeled learning; Precision Medicine; Rejection; Semi-supervised learning; Somatic variant calling; Supervised Machine Learning; Tandem Mass Spectrometry; Tumors; Vaccines; Whole Exome Sequencing; Peptide identification; Somatic variant calling; Positive-unlabeled learning; Tandem mass-spectrometry; Exome sequencing; Machine learning
• ISSN: 1471-2105; 1471-2105
• DOI: 10.1186/s12859-020-03813-x

Personalized cancer vaccines are emerging as one of the most promising approaches to immunotherapy of advanced cancers. However, only a small proportion of the neoepitopes generated by somatic DNA mutations in cancer cells lead to tumor rejection. Since it is impractical to experimentally assess all candidate neoepitopes prior to vaccination, developing accurate methods for predicting tumor-rejection mediating neoepitopes (TRMNs) is critical for enabling routine clinical use of cancer vaccines.
In this paper we introduce Positive-unlabeled Learning using AuTOml (PLATO), a general semi-supervised approach to improving accuracy of model-based classifiers. PLATO generates a set of high confidence positive calls by applying a stringent filter to model-based predictions, then rescores remaining candidates by using positive-unlabeled learning. To achieve robust performance on clinical samples with large patient-to-patient variation, PLATO further integrates AutoML hyper-parameter tuning, classification threshold selection based on spies, and support for bootstrapping.
Experimental results on real datasets demonstrate that PLATO has improved performance compared to model-based approaches for two key steps in TRMN prediction, namely somatic variant calling from exome sequencing data and peptide identification from MS/MS data.