Integrative Mechanistic Discovery Through Multi-Omics, AI-Driven Modeling, and Functional Validation

The field of molecular biosciences is undergoing rapid transformation driven by high-throughput technologies, multi-omics profiling, and advanced computational approaches. Large-scale datasets from genomics, transcriptomics, proteomics, metabolomics, and structural biology now enable unprecedented exploration of biological systems. Systems biology and computational modeling, supported by bioinformatics, artificial intelligence (AI), and machine learning (ML), are essential for converting these data into biological insights. However, a major challenge remains: ensuring that data-driven discoveries are mechanistically interpretable and experimentally validated.Many large-scale analyses produce associations or predictive models whose biological relevance is uncertain without functional validation. While integrative approaches combining single-cell, spatial omics, and network analysis are advancing the field, translating these findings into causal mechanisms requires robust in silico, in vitro, or in vivo validation. Notably, recent AI-driven advances, including biomolecular structure prediction and interaction modeling, have significantly expanded discovery capabilities, yet their interpretability and experimental grounding remain critical challenges.This Research Topic aims to bridge computational discovery and mechanistic understanding by promoting studies that directly connect data-driven findings with experimental validation. Particular emphasis will be placed on improving the interpretability of AI- and ML-based models through mechanistic insight and validation. We encourage contributions that integrate computational and experimental strategies, emphasizing reproducibility, transparency, and cross-disciplinary approaches to enhance biological and translational relevance.Within the boundaries of integrative computational and experimental studies in molecular biosciences, this Research Topic welcomes contributions that directly connect data analysis with functional and mechanistic validation. Topics include, but are not limited to: • Multi-omics integration and mechanistic network analysis • Computational modeling coupled with biochemical or biophysical validation • Systems biology approaches to elucidate molecular mechanisms • Single-cell and spatial omics linked to functional assays • Discovery and validation of molecular mechanisms in health and disease contexts (e.g., metabolic, immune, neurological, genetic disorders) • Data-driven identification and experimental validation of biomarkers and therapeutic targets • Methodologies for improving reproducibility and transparency in data-driven molecular biosciences • AI- and ML-based approaches for mechanistic discovery, including explainable and interpretable models • Data-driven algorithms for biomolecular structure prediction and their experimental or functional validation (e.g., protein structure, interactions, and dynamics)AppendixWe welcome Original Research Articles, Reviews, Methods, and Hypothesis and Theory papers. Submissions based primarily on computational analyses must include appropriate validation using independent datasets or experimental approaches. Purely descriptive studies, unvalidated predictions, or works lacking mechanistic insight are outside the scope. Topic editors: Qinan Yin, Yan Ma, Yonglan Liu Source: https://www.frontiersin.org/research-topics/80041/integrative-mechanistic-discovery-through-multi-omics-ai-driven-modeling-and-functional-validation