Abstract
Myogenesis, the process of muscle formation and regeneration, involves substantial alterations in gene expression. While alternative splicing plays a crucial role in generating proteomic diversity during development and disease, its specific contributions to human muscle differentiation have not been systematically explored. Here, we examined altered mRNA splicing during myogenesis in two human myoblast cell lines using a hybrid transcriptomic approach that combines short-read (Illumina) and long-read (Nanopore) RNA-seq analyses. We identified 13,853 new significant splicing isoforms (60,582 total), with RNAs increasing and decreasing in abundance between days 0 and 3 (3,771 and 3,649, respectively), and between days 3 and 5 (1,302 and 1,109, respectively). We identified 1,937 significant differential transcript usage events (DTUs), implicating pathways relevant for muscle regulation. These findings were validated using RT-qPCR analysis and across mouse and human models, including clinical samples from peripheral artery disease patients. Artificial Intelligence algorithms predicted 595 myogenesis-associated, high-confidence, novel protein-coding splicing isoforms. This study uncovers splicing-regulated mechanisms in muscle development and pathologies, establishing an integrative framework for studying mRNA processing, essential for future muscle biology intervention studies.
Supplementary Information:
The online version contains supplementary material available at 10.1038/s41598-025-16523-2.
Keywords:
AI-neural network; In-vitro muscle model; Myogenesis; Splicing; mRNA.