deep generative models, de novo molecule generation | INTRODUCTIONDe novo design of new molecules and analysis of their structure and properties is an important issue in computational molecular science. In the last few years, new approaches based on artificial intelligence (AI), especially deep learning models, have *These authors contributed equally to this study.

A-to-I RNA editing can contribute to the transcriptomic and proteomic diversity of many diseases including cancer. It has been reported that peptides generated from RNA editing could be naturally presented by human leukocyte antigen (HLA) molecules and elicit CD8+ T cell activation. However, a systematical characterization of A-to-I RNA editing neoantigens in cancer is still lacking. Here, an integrated RNA-editing based neoantigen identification pipeline PREP (Prioritizing of RNA Editing-based Peptides) was presented. A comprehensive RNA editing neoantigen profile analysis on 12 cancer types from The Cancer Genome Atlas (TCGA) cohorts was performed. PREP was also applied to 14 ovarian tumor samples and two clinical melanoma cohorts treated with immunotherapy. We finally proposed an RNA editing neoantigen immunogenicity score scheme, i.e. REscore, which takes RNA editing level and infiltrating immune cell population into consideration. We reported variant peptide from protein IFI30 in breast cancer which was confirmed expressed and presented in two samples with mass spectrometry data support. We showed that RNA editing neoantigen could be identified from RNA-seq data and could be validated with mass spectrometry data in ovarian tumor samples. Furthermore, we characterized the RNA editing neoantigen profile of clinical melanoma cohorts treated with immunotherapy. Finally, REscore showed significant associations with improved overall survival in melanoma cohorts treated with immunotherapy. These findings provided novel insights of cancer biomarker and enhance our understanding of neoantigen derived from A-to-I RNA editing as well as more types of candidates for personalized cancer vaccines design in the context of cancer immunotherapy.

No abstract

Next-generation sequencing has allowed identification of millions of somatic mutations in human cancer cells. A key challenge in interpreting cancer genomes is to distinguish drivers of cancer development among available genetic mutations. To address this issue, we present the first web-based application, consensus cancer driver gene caller (C3), to identify the consensus driver genes using six different complementary strategies, i.e., frequency-based, machine learning-based, functional bias-based, clustering-based, statistics model-based, and network-based strategies. This application allows users to specify customized operations when calling driver genes, and provides solid statistical evaluations and interpretable visualizations on the integration results. C3 is implemented in Python and is freely available for public use at http://drivergene.rwebox.com/c3.

No abstract