Optibrium demonstrates superior molecular docking method for small molecules and macrocycles

CAMBRIDGE, UK, 22 October 2024 – Optibrium, a leading developer of software and AI solutions for molecular design today announced the publication of a peer-reviewed study in the Journal of Computer-Aided Molecular Design, ‘Structure-Based Pose Prediction: Non-cognate Docking Extended to Macrocylic Ligands’. The paper demonstrates that Surflex-Dock can accurately predict the binding conformation and orientation (pose) of unknown (non-cognate) ligands, including macrocycles.

Molecular docking is a relatively easy process for re-docking known (cognate) ligands of proteins, exhibiting very high success rates. However, in any real-world application, it is unknown binding partners that are under investigation, such as predicting the binding mode of a candidate ligand during a drug design process. Predicting binding interactions in these scenarios is referred to as non-cognate molecular docking, which typically has a much lower success rate. Non-cognate docking is particularly challenging for macrocyclic compounds given their size and flexibility.

Using a data set of ~1000 ligands, the study compared different molecular docking methods to predict the non-cognate binding of non-macrocyclic and macrocyclic ligands. Using the Surflex-Dock method, correct poses among the two top-ranked predictions were identified 80% of the time, a success rate far exceeding those observed for AutoDock Vina and Gnina, particularly for macrocyclic ligands. In fact, the non-cognate macrocycle docking performance of Optibrium’s method rivals other reports of cognate non-macrocycle small molecule docking.

In addition to demonstrating the outstanding performance of Surflex-Dock for complex docking scenarios, the paper provides a benchmark for the community to determine the success of other methods on real-world molecular design applications.

Ann Cleves, VP of Application Science, BioPharmics Division, Optibrium, said: 

This paper reports, for the first time, highly accurate results for macrocyclic ligand conformation and orientation prediction which match the accuracy shown for diverse non-macrocyclic ligands. Crucially, the results characterising Surflex-Dock’s performance were achieved with ensembles of macrocycle-naive protein structures. Setting a new benchmark for non-cognate docking, our research underlines how Surflex-Dock is leading the field for chemists developing both macrocyclic and small molecule modalities and validates the use of Optibrium’s BioPharmics platform for developing macrocyclic therapeutics.