Protein Structure Prediction
Overview
Protein structure prediction aims to infer how amino-acid sequences form three-dimensional structures, which helps explain protein function and interactions. Approaches range from evolutionary information in multiple-sequence alignments and protein language models to geometric neural networks and generative methods such as diffusion; experimental restraints from cryo-EM or cross-linking can further guide predictions. Evaluation spans structure-accuracy benchmarks such as CAMEO and tests of complexes, binding, and physical plausibility. Models now tackle large assemblies and multiple conformations, but balancing accuracy, inference cost, reliable evaluation, and realistic dynamic behavior remains challenging.
Latest papers
47 papers9 papers in the last four weeks, level with the four weeks before. 0.1% of all new papers.