Receptor.AI introduces ArtiDock v2.0, the leading model for AI docking, delivering rapid and accurate predictions of ligand binding poses within protein pockets.
A detailed evaluation places ArtiDock against top AI docking methods and traditional programs like Vina, Gold, and Schrodinger's Glide.
ArtiDock surpasses AlphaFold-latest, the previous front-runner in predicting protein-ligand complexes, and exceeds classical docking algorithms in precision, significantly enhancing throughput.
Methodology
Benchmark Datasets Overview:
PoseBusters v1 Dataset: Specifically tailored to assess the quality of AI docking algorithms, focusing on their precision and reliability.
PoseBusters v3 Dataset: Aimed at further challenging AI docking models through more complex screening scenarios, testing their robustness and adaptability.
Performance Metrics Overview:
RMSD with Experimental Structures: Measures the accuracy of predicted structures against actual experimental data.
PB-Valid Set of Metrics: Evaluates not only the precision but also checks for steric clashes and the quality of the ligand conformers.
Results
ArtiDock outperforms all competitors on the PoseBusters v3 dataset, including Isomorphic Labs' latest AlphaFold.
ArtiDock notably faster, with physics-based docking algorithms and the Alpha-Fold model showing 20-600 times lower throughput compared to ArtiDock.
It is excels across all RSMD thresholds, ensuring accurate binding poses and identification of relevant chemical structures of ligands.
ArtiDock has the best precision and speed within methods providing relevant binding predictons.
Comparison of prediction accuracy among various docking methods
.png)
