Integration of structural study and machine learning to elucidate the RNA-SFs interaction atlas in eukaryotic cells

Abstract

Alternative splicing (AS) occupies a central position in plant growth and development, stress response, and animal growth and disease processes. Mutations in SF (splicing factor) trigger aberrant AS activities that disrupt these fine biological processes. Although cryo electron microscopy (cryoEM) technology has successfully revealed the fine structure of multiple spliceosomes, the dynamic and complex network of RNA-SFs remains to be fully resolved. This review summarizes the binding patterns of RNA and SFs through machine learning's powerful computational capabilities, the deep structural analysis using cryoEM, and experimental validation of RNA protein binding. Connect RNA protein interaction experiments, high-resolution imaging capabilities of cryoEM, and powerful analytical capabilities of machine learning to jointly construct a detailed RNA-SFs interaction map, forming a powerful toolkit. These knowledge help us better understand the complexity and working mechanisms of biological systems. This article not only has profound significance in revealing the molecular mechanisms of diseases and developing multi-target efficient drugs but also provides in-depth insights into molecular breeding and plant resistance enhancement.