Search

Summary: Self-supervised deep language modeling has shown unprecedented success across natural language tasks, and has recently been repurposed to biological sequences. However, existing models and pretraining methods are designed and optimized for text analysis. We introduce ProteinBERT, a deep language model specifically designed for proteins. Our pretraining scheme combines language modeling with a novel task of Gene Ontology (GO) annotation prediction. We introduce novel architectural elements that make the model highly efficient and flexible to long sequences. The architecture of ProteinBERT consists of both local and global representations, allowing end-to-end processing of these types of inputs and outputs. ProteinBERT obtains near state-of-the-art performance, and sometimes exceeds it, on multiple benchmarks covering diverse protein properties (including protein structure, post-translational modifications and biophysical attributes), despite using a far smaller and faster model than competing deep-learning methods. Overall, ProteinBERT provides an efficient framework for rapidly training protein predictors, even with limited labeled data., Availability and Implementation: Code and pretrained model weights are available at https://github.com/nadavbra/protein_bert., Supplementary Information: Supplementary data are available at Bioinformatics online., (© The Author(s) 2022. Published by Oxford University Press.)

Human genetic variation in coding regions is fundamental to the study of protein structure and function. Most methods for interpreting missense variants consider substitution measures derived from homologous proteins across different species. In this study, we introduce human-specific amino acid (AA) substitution matrices that are based on genetic variations in the modern human population. We analyzed the frequencies of >4.8M single nucleotide variants (SNVs) at codon and AA resolution and compiled human-centric substitution matrices that are fundamentally different from classic cross-species matrices (e.g. BLOSUM, PAM). Our matrices are asymmetric, with some AA replacements showing significant directional preference. Moreover, these AA matrices are only partly predicted by nucleotide substitution rates. We further test the utility of our matrices in exposing functional signals of experimentally-validated protein annotations. A significant reduction in AA transition frequencies was observed across nine post-translational modification (PTM) types and four ion-binding sites. Our results propose a purifying selection signal in the human proteome across a diverse set of functional protein annotations and provide an empirical baseline for interpreting human genetic variation in coding regions., (© The Author(s) 2021. Published by Oxford University Press on behalf of NAR Genomics and Bioinformatics.)

Compiling the catalogue of genes actively involved in cancer is an ongoing endeavor, with profound implications to the understanding and treatment of the disease. An abundance of computational methods have been developed to screening the genome for candidate driver genes based on genomic data of somatic mutations in tumors. Existing methods make many implicit and explicit assumptions about the distribution of random mutations. We present FABRIC, a new framework for quantifying the selection of genes in cancer by assessing the effects of de-novo somatic mutations on protein-coding genes. Using a machine-learning model, we quantified the functional effects of ∼3M somatic mutations extracted from over 10 000 human cancerous samples, and compared them against the effects of all possible single-nucleotide mutations in the coding human genome. We detected 593 protein-coding genes showing statistically significant bias towards harmful mutations. These genes, discovered without any prior knowledge, show an overwhelming overlap with known cancer genes, but also include many overlooked genes. FABRIC is designed to avoid false discoveries by comparing each gene to its own background model using rigorous statistics, making minimal assumptions about the distribution of random somatic mutations. The framework is an open-source project with a simple command-line interface., (© The Author(s) 2019. Published by Oxford University Press on behalf of Nucleic Acids Research.)