Predicting molecular mechanisms of hereditary diseases by using their tissue‐selective manifestation

• Published in: Molecular systems biology Vol. 19; no. 8; pp. e11407 - n/a
• Main Authors: Simonovsky, Eyal, Sharon, Moran, Ziv, Maya, Mauer, Omry, Hekselman, Idan, Jubran, Juman, Vinogradov, Ekaterina, Argov, Chanan M, Basha, Omer, Kerber, Lior, Yogev, Yuval, Segrè, Ayellet V, Im, Hae Kyung, Birk, Ohad, Rokach, Lior, Yeger‐Lotem, Esti
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 08.08.2023; EMBO Press; John Wiley and Sons Inc; Springer Nature
• Subjects: Consortia; data integration; Datasets; Decision trees; Deep learning; Disease; EMBO10; EMBO16; EMBO22; Gene expression; Genes; Genetic disorders; genomic medicine; Hereditary diseases; Humans; Learning algorithms; Life Sciences; Machine Learning; Molecular modelling; Mutation; omics; Ontology; Physiology; Proteins; Questions; Risk assessment; Selectivity; Systems Biology; tissue selectivity; Tissues; genomic medicine; omics; machine learning; tissue selectivity; data integration
• ISSN: 1744-4292; 1744-4292
• DOI: 10.15252/msb.202211407

How do aberrations in widely expressed genes lead to tissue‐selective hereditary diseases? Previous attempts to answer this question were limited to testing a few candidate mechanisms. To answer this question at a larger scale, we developed “Tissue Risk Assessment of Causality by Expression” (TRACE), a machine learning approach to predict genes that underlie tissue‐selective diseases and selectivity‐related features. TRACE utilized 4,744 biologically interpretable tissue‐specific gene features that were inferred from heterogeneous omics datasets. Application of TRACE to 1,031 disease genes uncovered known and novel selectivity‐related features, the most common of which was previously overlooked. Next, we created a catalog of tissue‐associated risks for 18,927 protein‐coding genes ( https://netbio.bgu.ac.il/trace/ ). As proof‐of‐concept, we prioritized candidate disease genes identified in 48 rare‐disease patients. TRACE ranked the verified disease gene among the patient's candidate genes significantly better than gene prioritization methods that rank by gene constraint or tissue expression. Thus, tissue selectivity combined with machine learning enhances genetic and clinical understanding of hereditary diseases. Synopsis An interpretable machine‐learning framework predicts disease genes for tissue‐selective hereditary diseases. The framework highlights known and novel tissue‐selectivity features and enhances genetic diagnosis of rare‐disease patients. An interpretable machine‐learning (ML) framework uses thousands of gene features to predict tissue‐associated disease genes. ML models highlight known and novel tissue‐selectivity mechanisms. An online catalogue of tissue‐associated risks for 18,927 protein‐coding genes in eight tissues is presented. The framework and catalogue enhance genetic diagnosis of rare‐disease patients. Graphical Abstract An interpretable machine‐learning framework predicts disease genes for tissue‐selective hereditary diseases. The framework highlights known and novel tissue‐selectivity features and enhances genetic diagnosis of rare‐disease patients.