ProModEv (Protein Model Evaluation portal) is a platform developed for benchmarking and evaluating Artificial Intelligence (AI) and Machine Learning (ML) tools used in protein structure prediction, including AlphaFold2, AlphaFold3, and ESMFold. It features an interactive, user-friendly dashboard that offers clear visualizations of performance metrics and comparative data, facilitating seamless exploration and analysis.
The primary aim of ProModEv is to systematically assess the accuracy of these AI/ML-based methods relative to experimentally determined structures sourced from the Protein Data Bank (PDB). The portal underscores the prediction accuracy of these tools for both monomeric and oligomeric proteins, aiding researchers in identifying their strengths and limitations.
Currently, ProModEv hosts evaluation results for challenging datasets comprising 1666 monomeric and 994 dimeric proteins deposited in the PDB between 2022 and 2024. Currently, efforts are going in the direction of expanding its coverage on a weekly basis by incorporating all newly deposited structures in the PDB.
