DynamicBind: Predicting ligand-specific protein-ligand complex structure with a deep equivariant generative model

Abstract

While significant advances have been made in predicting static protein structures, the inherent dynamics of proteins, modulated by ligands, are crucial for understanding protein function and facilitating drug discovery. Traditional docking methods, frequently used in studying protein-ligand interactions, typically treat proteins as rigid. While molecular dynamics simulations can propose appropriate protein conformations, they’re computationally demanding due to rare transitions between biologically relevant equilibrium states. In this study, we present DynamicBind, a novel method that employs equivariant geometric diffusion networks to construct a smooth energy landscape, promoting efficient transitions between different equilibrium states. DynamicBind accurately recovers ligand-specific conformations from unbound protein structures without the need for holo-structures or extensive sampling. Remarkably, it demonstrates state-of-the-art performance in docking and virtual screening benchmarks. Our experiments reveal that DynamicBind can accommodate a wide range of large protein conformational changes and identify novel cryptic pockets in unseen protein targets. As a result, DynamicBind shows potential in accelerating the development of small molecules for previously undruggable targets and expanding the horizons of computational drug discovery.

Advancing Ligand Docking through Deep Learning: Challenges and Prospects in Virtual Screening

“…However, when dealing with scenarios where only the unbound protein structure (apo) is available or when protein structures are predicted using AlphaFold2, a significant drop in performance is observed for KarmaDock and other DLLD methods. Since then, a series of flexible docking methodologies , and several de novo LD methods − (Figure D) have been developed. These methods are able to generate ligand binding poses and modify protein conformations or generate the conformations for proteins and ligands from sequences at the cost of speed.…”

“…Currently, a few DL strategies, ,,, such as KarmaDock, have successfully addressed these aspects, demonstrating enhanced docking and screening power on CASF-2016. Specifically, KarmaDock stands out not only for its docking module designed to predict binding poses but also for its integration of an MDN module tailored to PL distances, thereby facilitating a more accurate assessment of binding strength.…”

“…From a flexibility standpoint, the initial focus was on semiflexible DLLD methods, deemed sufficient for most VS scenarios and aligned with the design of most traditional LD programs. Currently, there is a growing interest in developing flexible docking methodologies and de novo methods that generate PL complex conformations directly from sequences. − This interest is driven by the scenarios where only unbound protein structures (apo structures), available through databases like PDB or predicted by tools like AlphaFold2, RosettaFold, and ESMFold, are accessible. These unbound structures may significantly differ from their bound counterparts (holo structures), and semiflexible DLLD methods, often trained on data sets comprising holo structures, have limited generalizability to apo structures.…”

“…The majority of existing DLLD methodologies ,− ,, including KarmaDock suffer from inadequate generalization capabilities, resulting in excellent performance on samples resembling the training set but decreased effectiveness or the generation of physically implausible conformations on out-of-distribution samples. Therefore, enhancing the generalization capacity of the DLLD approaches is crucial.…”

Diffusion models in protein structure and docking

“…Unfortunately, there is no common universally adopted definition of pocket, making empirical direct comparisons between these approaches very hard. For instance, different methods 88 use boxes or spheres of various sizes around the true ligand pose to define the pocket, while others 85 consider the pocket as the set of residues of the proteins interacting with the ligand. 2.…”