Unveiling MiRNA-124 as a biomarker in hypertrophic cardiomyopathy: An innovative approach using machine learning and intelligent data analysis.

ABSTRACT

BACKGROUND: Hypertrophic cardiomyopathy (HCM) is a widespread hereditary cardiac pathology characterized by thickened heart walls and rearrangement of cardiomyocytes. Despite extensive research, the mechanisms underlying HCM development remain poorly understood, impeding the development of effective therapeutic and diagnostic strategies. Recent studies have suggested a polygenic nature of HCM development alongside monogenic forms. Transcriptomic profiling is a valuable tool for investigating such diseases. In this study, we propose a novel approach to study regulatory microRNAs (miRNAs) in the context of HCM, utilizing state-of-the-art data analysis tools. METHODS AND RESULTS: Our method involves applying the Monte Carlo simulation and machine learning algorithm to transcriptomic data to generate high-capacity classifiers for HCM. From these classifiers, we extract key genes crucial for their performance, resulting in the identification of 16 key genes. Subsequently, we narrow down the pool of miRNAs by selecting those that may target the greatest number of key genes within the best models. We particularly focused on miR-124-3p, which we validated to have an association with HCM on an independent dataset. Subsequent investigation of its function revealed involvement of miR-124-3p in the RhoA signaling pathway. CONCLUSIONS: In this study we propose a new approach to analyze transcriptomic data to search for microRNAs associated with a disease. Using this approach for transcriptomic profiling data of patients with HCM, we identified miR-124-3p as a potential regulator of the RhoA signaling pathway in the pathogenesis of HCM.