Single amino acid substitutions can profoundly affect protein folding, dynamics, and function, leading to potential pathological consequences. The ability to discern between benign and pathogenic substitutions is pivotal for therapeutic interventions and research directions. Given the limitations in experimental examination of these variants, AlphaMissense has emerged as a promising predictor of the pathogenicity of single nucleotide polymorphism variants. In our study, we assessed the efficacy of AlphaMissense across several protein groups, such as mitochondrial, housekeeping, transmembrane proteins, and specific proteins like CFTR, using ClinVar data for validation. Our comprehensive evaluation showed that AlphaMissense delivers outstanding performance, with MCC scores predominantly between 0.6 and 0.74. We observed low performance on the CFTR and disordered, membrane-interacting MemMoRF datasets. However, an enhanced performance with CFTR was shown when benchmarked against the CFTR2 database. Our results also emphasize that quality of AlphaFold9s predictions can seriously influence AlphaMissense predictions. Most importantly, AlphaMissense9s consistent capability in predicting pathogenicity across diverse protein groups, spanning both transmembrane and soluble domains was found. Moreover, the prediction of likely-pathogenic labels for IBS and CFTR coupling helix residues emphasizes AlphaMissense9s potential as a tool for pinpointing functionally significant sites. Additionally, to make AlphaMissense predictions more accessible, we have introduced a user-friendly web resource (https://alphamissense.hegelab.org) to enhance the utility of this valuable tool. Our insights into AlphaMissense9s capability, along with this online resource, underscore its potential to significantly aid both research and clinical applications.

Lightway access to AlphaMissense data that demonstrates a balanced performance of this missense mutation predictor