A Flexible Approach to Induced Fit Docking

Abstract

We present Fleksy, a new approach to consider both ligand and receptor flexibility in small molecule docking. Pivotal to our method is the use of a receptor ensemble to describe protein flexibility. To construct these ensembles, we use a backbone-dependent rotamer library and implement the concept of interaction sampling. The latter allows the evaluation of different orientations of ambivalent interaction partners. The docking stage consists of an ensemble-based soft-docking experiment using FlexX-Ensemble, followed by an effective flexible receptor-ligand complex optimization using Yasara. Fleksy produces a set of receptor-ligand complexes ranked using a consensus scoring function combining docking scores and force field energies. Averaged over three cross-docking datasets, containing 35 different receptor-ligand complexes in total, Fleksy reproduces the observed binding mode within 2.0 Å for 78% of the complexes. This compares favorably to the rigid receptor FlexX program, which on average reaches a success rate of 44% for these datasets.

Thermomyces lanuginosus Lipase with Closed Lid Catalyzes Elimination of Acetic Acid from 11‐Acetyl‐Prostaglandin E2

“…The final results of the work indicate that this simplification was valid. Molecular mechanics and MD calculations were performed with the software package YASARA [50,51] in which AMBER99 force field was used. MD simulation methodology was used to generate TLL structures solvated in benzene and in methanol.…”

Understanding the challenges of protein flexibility in drug design

“…Some methods explicitly perform a similar connection, mixing ensemble docking and induced-fit docking [67], [132]. Other examples are approaches that combine ensemble docking and molecular-relaxation docking [118], [133]. In this case, each binding pose of the protein-ligand complex is refined using an optimization algorithm.…”

Protein Flexibility in In Silico Screening

“…Using energy minimization instead of MD simulations helps to overcome this problem, although a more thorough sampling of the protein conformational space is now required in the initial docking step. This philosophy is followed by the Fleksy approach [150], which uses a FlexE-based ensemble docking, followed by an effective Yasara-based complex optimization. Similarly, the previously published RosettaLigand method [151], which only allowed for protein side-chain degrees of freedom so far, has been extended by a stringent gradient-based minimization in the final step [152], now allowing side-chain and backbone torsions of the receptor to vary.…”