S Kim6, H S Kim2, E Kim1, M G Lee3, E-C Shin4, S Paik5, S Kim6. 1. Severance Biomedical Science Institute. 2. Department of Pharmacology, Pharmacogenomic Research Center for Membrane Transporters, Brain Korea 21 PLUS Project for Medical Sciences; Division of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul. 3. Department of Pharmacology, Pharmacogenomic Research Center for Membrane Transporters, Brain Korea 21 PLUS Project for Medical Sciences. 4. Graduate School of Medical Science and Engineering, KAIST, Daejeon, Korea. 5. Severance Biomedical Science Institute; Division of Medical Oncology, Department of Internal Medicine, Yonsei Cancer Center, Yonsei University College of Medicine, Seoul. 6. Severance Biomedical Science Institute. Electronic address: swkim@yuhs.ac.
Abstract
Background: Tumor-specific mutations form novel immunogenic peptides called neoantigens. Neoantigens can be used as a biomarker predicting patient response to cancer immunotherapy. Although a predicted binding affinity (IC50) between peptide and major histocompatibility complex class I is currently used for neoantigen prediction, large number of false-positives exist. Materials and methods: We developed Neopepsee, a machine-learning-based neoantigen prediction program for next-generation sequencing data. With raw RNA-seq data and a list of somatic mutations, Neopepsee automatically extracts mutated peptide sequences and gene expression levels. We tested 14 immunogenicity features to construct a machine-learning classifier and compared with the conventional methods based on IC50 regarding sensitivity and specificity. We tested Neopepsee on independent datasets from melanoma, leukemia, and stomach cancer. Results: Nine of the 14 immunogenicity features that are informative and inter-independent were used to construct the machine-learning classifiers. Neopepsee provides a rich annotation of candidate peptides with 87 immunogenicity-related values, including IC50, expression levels of neopeptides and immune regulatory genes (e.g. PD1, PD-L1), matched epitope sequences, and a three-level (high, medium, and low) call for neoantigen probability. Compared with the conventional methods, the performance was improved in sensitivity and especially two- to threefold in the specificity. Tests with validated datasets and independently proven neoantigens confirmed the improved performance in melanoma and chronic lymphocytic leukemia. Additionally, we found sequence similarity in proteins to known pathogenic epitopes to be a novel feature in classification. Application of Neopepsee to 224 public stomach adenocarcinoma datasets predicted ∼7 neoantigens per patient, the burden of which was correlated with patient prognosis. Conclusions: Neopepsee can detect neoantigen candidates with less false positives and be used to determine the prognosis of the patient. We expect that retrieval of neoantigen sequences with Neopepsee will help advance research on next-generation cancer immunotherapies, predictive biomarkers, and personalized cancer vaccines.
Background: Tumor-specific mutations form novel immunogenic peptides called neoantigens. Neoantigens can be used as a biomarker predicting patient response to cancer immunotherapy. Although a predicted binding affinity (IC50) between peptide and major histocompatibility complex class I is currently used for neoantigen prediction, large number of false-positives exist. Materials and methods: We developed Neopepsee, a machine-learning-based neoantigen prediction program for next-generation sequencing data. With raw RNA-seq data and a list of somatic mutations, Neopepsee automatically extracts mutated peptide sequences and gene expression levels. We tested 14 immunogenicity features to construct a machine-learning classifier and compared with the conventional methods based on IC50 regarding sensitivity and specificity. We tested Neopepsee on independent datasets from melanoma, leukemia, and stomach cancer. Results: Nine of the 14 immunogenicity features that are informative and inter-independent were used to construct the machine-learning classifiers. Neopepsee provides a rich annotation of candidate peptides with 87 immunogenicity-related values, including IC50, expression levels of neopeptides and immune regulatory genes (e.g. PD1, PD-L1), matched epitope sequences, and a three-level (high, medium, and low) call for neoantigen probability. Compared with the conventional methods, the performance was improved in sensitivity and especially two- to threefold in the specificity. Tests with validated datasets and independently proven neoantigens confirmed the improved performance in melanoma and chronic lymphocytic leukemia. Additionally, we found sequence similarity in proteins to known pathogenic epitopes to be a novel feature in classification. Application of Neopepsee to 224 public stomach adenocarcinoma datasets predicted ∼7 neoantigens per patient, the burden of which was correlated with patient prognosis. Conclusions: Neopepsee can detect neoantigen candidates with less false positives and be used to determine the prognosis of the patient. We expect that retrieval of neoantigen sequences with Neopepsee will help advance research on next-generation cancer immunotherapies, predictive biomarkers, and personalized cancer vaccines.
Authors: Shutao Mei; Fuyi Li; André Leier; Tatiana T Marquez-Lago; Kailin Giam; Nathan P Croft; Tatsuya Akutsu; A Ian Smith; Jian Li; Jamie Rossjohn; Anthony W Purcell; Jiangning Song Journal: Brief Bioinform Date: 2020-07-15 Impact factor: 11.622
Authors: Christof C Smith; Shengjie Chai; Amber R Washington; Samuel J Lee; Elisa Landoni; Kevin Field; Jason Garness; Lisa M Bixby; Sara R Selitsky; Joel S Parker; Barbara Savoldo; Jonathan S Serody; Benjamin G Vincent Journal: Cancer Immunol Res Date: 2019-09-12 Impact factor: 11.151
Authors: Christof C Smith; Sara R Selitsky; Shengjie Chai; Paul M Armistead; Benjamin G Vincent; Jonathan S Serody Journal: Nat Rev Cancer Date: 2019-07-05 Impact factor: 60.716