Machine learning analysis of ARVC informed by sodium channel protein-based interactome networks

Area of Science:

• Cardiovascular Research
• Pharmacology
• Computational Biology

Background:

• Arrhythmogenic right ventricular cardiomyopathy (ARVC) is an inherited condition linked to sodium channel dysfunction.
• Current clinical management strategies for ARVC are limited, necessitating the discovery of new therapeutic agents.
• Developing novel compounds is critical for advancing ARVC drug development and improving patient outcomes.

Purpose of the Study:

• To identify and validate novel chemical compounds with therapeutic potential for arrhythmogenic right ventricular cardiomyopathy (ARVC).
• To leverage machine learning and organoid models for efficient drug discovery in inherited cardiac disorders.

Main Methods:

• Utilized scRNA-seq data from ARVC rat models to construct machine learning (ML) models based on protein-protein interaction (PPI) networks.
• Predicted binding affinity (BA) of compounds to key proteins implicated in ARVC.
• Validated ML predictions through in vitro experiments using cardiac organoids, including Western blotting, ELISA, MEA, and Masson staining.

Main Results:

• Identified SCN5A as the most significantly affected sodium channel protein in arrhythmogenic right ventricular cardiomyopathy (ARVC).
• Machine learning models accurately predicted high binding affinity between Kaempferol and SCN5A.
• In vitro studies confirmed that Kaempferol demonstrates significant therapeutic effects in ARVC models.

Conclusions:

• This study introduces an innovative platform integrating ML and organoid validation for identifying ARVC therapeutic compounds.
• The findings highlight Kaempferol as a promising lead compound for the treatment of arrhythmogenic right ventricular cardiomyopathy (ARVC).
• This approach offers a valuable strategy for addressing public health challenges associated with ARVC and has broad applicability.