Computational Glycoscience: From Physics-Based Modeling to Machine Learning

The structural complexity of glycans, their branching, context-dependence, and dynamic remodelling, has long made them the most data-poor class of biomolecules in systems biology. Recent advances in machine learning, molecular modelling, and glyco-informatics are beginning to close this “glycan gap”, enabling researchers to decode glycan structures, functions, and interactions at a scale and resolution that was previously out of reach.This Research Topic invites contributions that advance the development and application of computational methods in glycoscience. Topics of interest span the full methodological spectrum: algorithms for glycan structure prediction and annotation; tools for glycoproteomics data analysis; glycan–protein docking and molecular dynamics simulations; network and pathway modelling of glycosylation; and AI-driven approaches to glycomics dataset interpretation. We also welcome work that bridges computational glycoscience with translational or clinical applications, including glycan-based biomarker discovery and the rational design of glycoengineered therapeutics.We particularly welcome submissions that:- Integrate multi-omics data to dissect the regulation of glycosylation at the systems level- Develop interoperable, FAIR-compliant glyco-databases, ontologies, and data standards- Enable predictive modelling of glycan function across cells, tissues, and organisms- Present novel software tools, benchmarks, or reproducible computational workflowsOriginal research, methods papers, software tool descriptions, systematic benchmarks, and application case studies are all in scope. Topic editors: Matthew P. DeLisa, Cleo Kontoravdi, Pavlos Kotidis Source: https://www.frontiersin.org/research-topics/81939/computational-glycoscience-from-physics-based-modeling-to-machine-learning