Narrow Results

Elucidating the consequences of mutations on protein function is important for identifying the functionally critical residues and designing proteins with enhanced function. Currently, several databases accumulate information on protein functions for experimentally known variants. On the other hand, prediction tools have been developed to reveal the effects of mutations on protein function. In this chapter, the advancements in experimental methods from traditional site-directed mutagenesis to recent deep mutational scanning experiments are discussed. In addition, we explain variant effect prediction methods ranging from simple amino acid substitution probabilities to state-of-the-art zero-shot language models, along with potential applications. Further, recent efforts such as the “Atlas of variant effects alliance” to accelerate research in variant effect prediction and the Critical Assessment of Genome Interpretation (CAGI) experiment for assessing the performance of the different variant effect predictors are outlined.