Combining Transition Path Sampling with Data-Driven Collective Variables through a Reactivity-Biased Shooting Algorithm.

Rare event sampling is a central problem in modern computational chemistry research. Among the existing methods, transition path sampling (TPS) can generate unbiased representations of reaction processes. However, its efficiency depends on the ability to generate reactive trial paths, which in turn depends on the quality of the shooting algorithm used. We propose a new algorithm based on the shooting success rate, i.e., reactivity, measured as a function of a reduced set of collective variables (CVs). These variables are extracted with a machine learning approach directly from TPS simulations, using a multitask objective function. Iteratively, this workflow significantly improves the shooting efficiency without any prior knowledge of the process. In addition, the optimized CVs can be used with biased enhanced sampling methodologies to accurately reconstruct the free energy profiles. We tested the method on three different systems: a two-dimensional toy model, conformational transitions of alanine dipeptide, and hydrolysis of acetyl chloride in bulk water. In the latter, we integrated our workflow with an active learning scheme to learn a reactive machine learning-based potential, which allowed us to study the mechanism and free energy profile with an ab initio-like accuracy.

Comprehensive Evaluation of 10 Docking Programs on a Diverse Set of Protein-Cyclic Peptide Complexes.

Cyclic peptides have emerged as a highly promising class of therapeutic molecules owing to their favorable pharmacokinetic properties, including stability and permeability. Currently, many clinically approved cyclic peptides are derived from natural products or their derivatives, and the development of molecular docking techniques for cyclic peptide discovery holds great promise for expanding the applications and potential of this class of molecules. Given the availability of numerous docking programs, there is a pressing need for a systematic evaluation of their performance, specifically on protein-cyclic peptide systems. In this study, we constructed an extensive benchmark data set called CPSet, consisting of 493 protein-cyclic peptide complexes. Based on this data set, we conducted a comprehensive evaluation of 10 docking programs, including Rosetta, AutoDock CrankPep, and eight protein-small molecule docking programs (i.e., AutoDock, AudoDock Vina, Glide, GOLD, LeDock, rDock, MOE, and Surflex). The evaluation encompassed the assessment of the sampling power, docking power, and scoring power of these programs. The results revealed that all of the tested protein-small molecule docking programs successfully sampled the binding conformations when using the crystal conformations as the initial structures. Among them, rDock exhibited outstanding performance, achieving a remarkable 94.3% top-100 sampling success rate. However, few programs achieved successful predictions of the binding conformations using tLEaP-generated conformations as the initial structures. Within this scheme, AutoDock CrankPep yielded the highest top-100 sampling success rate of 29.6%. Rosetta's scoring function outperformed the others in selecting optimal conformations, resulting in an impressive top-1 docking success rate of 87.6%. Nevertheless, all the tested scoring functions displayed limited performance in predicting binding affinity, with MOE@Affinity dG exhibiting the highest Pearson's correlation coefficient of 0.378. It is therefore suggested to use an appropriate combination of different docking programs for given tasks in real applications. We expect that this work will offer valuable insights into selecting the appropriate docking programs for protein-cyclic peptide complexes.

Comprehensive, Open-Source, and Automated Workflow for Multisite λ-Dynamics in Lead Optimization.

Multisite λ-dynamics (MSLD) is a highly efficient binding free energy calculation method that samples multiple ligands in a single round by assigning different λ values to the alchemical part of each ligand. This method holds great promise for lead optimization (LO) in drug discovery. However, the complex data preparation and simulation process limits its widespread application in diverse protein-ligand systems. To address this challenge, we developed a comprehensive, open-source, and automated workflow for MSLD calculations based on the BLaDE dynamics engine. This workflow incorporates the Ligand Internal and Cartesian coordinate reconstruction-based alignment algorithm (LIC-align) and an optimized maximum common substructure (MCS) search algorithm to accurately generate MSLD multiple topologies with ideal perturbation patterns. Furthermore, our workflow is highly modularized, allowing straightforward integration and extension of various simulation techniques, and is highly accessible to nonexperts. This workflow was validated by calculating the relative binding free energies of large-scale congeneric ligands, many of which have large perturbing groups. The agreement between the calculations and experiments was excellent, with an average unsigned error of 1.08 ± 0.47 kcal/mol. More than 57.1% of the ligands had an error of less than 1.0 kcal/mol, and the perturbations of 6 targets were fully connected via the calculations, while those of 2 targets were connected via both calculations and experimental data. The Pearson correlation coefficient reached 0.88, indicating that the MSLD workflow provides accurate predictions that can guide lead optimization in drug discovery. We also examined the impact of single-site versus multisite perturbations, ligand grouping by perturbing group size, and the position of the anchor atom on the MSLD performance. By integrating our proposed LIC-align and optimized MCS search algorithm along with the coping strategies to handle challenging molecular substructures, our workflow can handle many realistic scenarios more reasonably than all previously published methods. Moreover, we observed that our MSLD workflow achieved similar accuracy to free energy perturbation (FEP) while improving computational efficiency by over 1 order of magnitude in speedup. These findings provide valuable insights and strategies for further MSLD development, making MSLD a competitive tool for lead optimization.

Small-Molecule Conformer Generators: Evaluation of Traditional Methods and AI Models on High-Quality Data Sets.

Small-molecule conformer generation (SMCG) is an extremely important task in both ligand- and structure-based computer-aided drug design, especially during the hit discovery phase. Recently, a multitude of artificial intelligence (AI) models tailored for SMCG have emerged. Despite developers typically furnishing performance evaluation data upon releasing their AI models, a comprehensive and equitable performance comparison between AI models and conventional methods is still lacking. In this study, we curated a new benchmarking data set comprising 3354 high-quality ligand bioactive conformations. Subsequently, we conducted a systematic assessment of the performance of four widely adopted traditional methods (i.e., ConfGenX, Conformator, OMEGA, and RDKit ETKDG) and five AI models (i.e., ConfGF, DMCG, GeoDiff, GeoMol, and torsional diffusion) in the tasks of reproducing bioactive and low-energy conformations of small molecules. In the former task, the AI models have no advantage, particularly with a maximum ensemble size of 1. Even the best-performing AI model GeoMol is still worse than any of the tested traditional methods. Conversely, in the latter task, the torsional diffusion model shows obvious advantages, surpassing the best-performing traditional method ConfGenX by 26.09 and 12.97% on the COV-R and COV-P metrics, respectively. Furthermore, the influence of force field-based fine-tuning on the quality of the generated conformers was also discussed. Finally, a user-friendly Web server called fastSMCG was developed to enable researchers to rapidly and flexibly generate small-molecule conformers using both traditional and AI methods. We anticipate that our work will offer valuable practical assistance to the scientific community in this field.

On the Dynamic Mechanism of Long-Flexible Fatty Acid Binding to Fatty Acid Binding Protein: Resolving the Long-Standing Debate.

Fatty acids (FAs) are one of the essential energy sources for physiological processes, and they play a vital role in regulating immune and inflammatory responses, promoting cell differentiation and apoptosis, and inhibiting tumor growth. These functions are carried out by FA binding proteins (FABPs) that recognize and transport FAs. Although the crystal structure of the FA-FABPs complex has long been characterized, the mechanism behind FA binding and dissociation from FABP remains unclear. This study employed conventional MD simulations and enhanced sampling technologies to investigate the atomic-scale complexes of heart fatty acid binding proteins and stearic acid (SA). The results revealed two primary pathways for the binding or dissociation of the flexible long-chain ligand, with the orientation of the SA carboxyl head during dissociation determining the chosen path. Conformational changes in the portal region of FABP during the ligand binding/unbinding were found to be trivial, and the overturn of the ″cap″ or the unfolding of the α2 helix was not required. This study resolves the long-standing debate on the binding mechanism of SA with the long-flexible tail to FABP, which significantly improves the understanding of the transport mechanism of FABPs and the development of related therapeutic agents.

TB-IECS: an accurate machine learning-based scoring function for virtual screening.

Machine learning-based scoring functions (MLSFs) have shown potential for improving virtual screening capabilities over classical scoring functions (SFs). Due to the high computational cost in the process of feature generation, the numbers of descriptors used in MLSFs and the characterization of protein-ligand interactions are always limited, which may affect the overall accuracy and efficiency. Here, we propose a new SF called TB-IECS (theory-based interaction energy component score), which combines energy terms from Smina and NNScore version 2, and utilizes the eXtreme Gradient Boosting (XGBoost) algorithm for model training. In this study, the energy terms decomposed from 15 traditional SFs were firstly categorized based on their formulas and physicochemical principles, and 324 feature combinations were generated accordingly. Five best feature combinations were selected for further evaluation of the model performance in regard to the selection of feature vectors with various length, interaction types and ML algorithms. The virtual screening power of TB-IECS was assessed on the datasets of DUD-E and LIT-PCBA, as well as seven target-specific datasets from the ChemDiv database. The results showed that TB-IECS outperformed classical SFs including Glide SP and Dock, and effectively balanced the efficiency and accuracy for practical virtual screening.

SDEGen: learning to evolve molecular conformations from thermodynamic noise for conformation generation.

Generation of representative conformations for small molecules is a fundamental task in cheminformatics and computer-aided drug discovery, but capturing the complex distribution of conformations that contains multiple low energy minima is still a great challenge. Deep generative modeling, aiming to learn complex data distributions, is a promising approach to tackle the conformation generation problem. Here, inspired by stochastic dynamics and recent advances in generative modeling, we developed SDEGen, a novel conformation generation model based on stochastic differential equations. Compared with existing conformation generation methods, it enjoys the following advantages: (1) high model capacity to capture multimodal conformation distribution, thereby searching for multiple low-energy conformations of a molecule quickly, (2) higher conformation generation efficiency, almost ten times faster than the state-of-the-art score-based model, ConfGF, and (3) a clear physical interpretation to learn how a molecule evolves in a stochastic dynamics system starting from noise and eventually relaxing to the conformation that falls in low energy minima. Extensive experiments demonstrate that SDEGen has surpassed existing methods in different tasks for conformation generation, interatomic distance distribution prediction, and thermodynamic property estimation, showing great potential for real-world applications.

Quasiclassical Trajectory Simulation as a Protocol to Build Locally Accurate Machine Learning Potentials.

Direct trajectory calculations have become increasingly popular in recent computational chemistry investigations. However, the exorbitant computational cost of ab initio trajectory calculations usually limits its application in mechanistic explorations. Recently, machine learning-based potential energy surface (ML-PES) provides a powerful strategy to circumvent the heavy computational cost and meanwhile maintain the required accuracy. Despite the appealing potential, constructing a robust ML-PES is still challenging since the training set of the PES should cover a broad enough configuration space. In this work, we demonstrate that when the concerned properties could be collected by the localized sampling of the configuration space, quasiclassical trajectory (QCT) calculations can be invoked to efficiently obtain locally accurate ML-PESs. We prove our concept with two model reactions: methyl migration of i-pentane cation and dimerization of cyclopentadiene. We found that the locally accurate ML-PESs are sufficiently robust for reproducing the static and dynamic features of the reactions, including the time-resolved free energy and entropy changes, and time gaps.

Sigmoid Accelerated Molecular Dynamics: An Efficient Enhanced Sampling Method for Biosystems.

Gaussian accelerated molecular dynamics (GaMD) is recognized as a popular enhanced sampling method for tackling long-standing challenges in biomolecular simulations. Inspired by GaMD, Sigmoid accelerated molecular dynamics (SaMD) is proposed in this work by adding a Sigmoid boost potential to improve the balance between the highest acceleration and accurate reweighting. Compared with GaMD, SaMD extends the accessible time scale and improves the computational efficiency as tested in three tasks. In the alanine dipeptide task, SaMD can produce the free energy landscape with better accuracy and efficiency. In the chignolin folding task, the estimated Gibbs free energy difference can converge to the experimental value ∼30% faster. In the protein-ligand binding task, the bound conformations are closer to the crystal structure with a minimal ligand root-mean-square deviation of 1.7 Å. The binding of the ligand XK263 to the HIV protease is reproduced by SaMD in ∼60% less simulation time.

Efficient and accurate large library ligand docking with KarmaDock.

Ligand docking is one of the core technologies in structure-based virtual screening for drug discovery. However, conventional docking tools and existing deep learning tools may suffer from limited performance in terms of speed, pose quality and binding affinity accuracy. Here we propose KarmaDock, a deep learning approach for ligand docking that integrates the functions of docking acceleration, binding pose generation and correction, and binding strength estimation. The three-stage model consists of the following components: (1) encoders for the protein and ligand to learn the representations of intramolecular interactions; (2) E(n) equivariant graph neural networks with self-attention to update the ligand pose based on both protein-ligand and intramolecular interactions, followed by post-processing to ensure chemically plausible structures; (3) a mixture density network for scoring the binding strength. KarmaDock was validated on four benchmark datasets and tested in a real-world virtual screening project that successfully identified experiment-validated active inhibitors of leukocyte tyrosine kinase (LTK).

Discovery of Novel GR Ligands toward Druggable GR Antagonist Conformations Identified by MD Simulations and Markov State Model Analysis.

Binding of different ligands to glucocorticoid receptor (GR) may induce different conformational changes and even trigger completely opposite biological functions. To understand the allosteric communication within the GR ligand binding domain, the folding pathway of helix 12 (H12) induced by the binding of the agonist dexamethasone (DEX), antagonist RU486, and modulator AZD9567 are explored by molecular dynamics simulations and Markov state model analysis. The ligands can regulate the volume of the activation function-2 through the residues Phe737 and Gln738. Without ligand or with agonist binding, H12 swings from inward to outward to visit different folding positions. However, the binding of RU486 or AZD9567 perturbs the structural state, and the passive antagonist state appears more stable. Structure-based virtual screening and in vitro bioassays are used to discover novel GR ligands that bias the conformation equilibria toward the passive antagonist state. HP-19 exhibits the best anti-inflammatory activity (IC50 = 0.041 ± 0.011 µm) in nuclear factor-kappa B signaling pathway, which is comparable to that of DEX. HP-19 also does not induce adverse effect-related transactivation functions of GR. The novel ligands discovered here may serve as promising starting points for the development of GR modulators.

Investigation of ECD conformational transition mechanism of GLP-1R by molecular dynamics simulations and Markov state model.

As a member of the class B G protein-coupled receptors (GPCRs), the glucagon-like peptide-1 (GLP-1) can regulate the blood glucose level by binding to the glucagon-like peptide-1 receptor (GLP-1R). Since the extracellular domain (ECD) of GLP-1R is considered as one of the binding sites of GLP-1, the open and closed states of ECD play an important role in the binding process of GLP-1. To investigate the transition path of GLP-1R ECD, the crystal structures of GLP-1R in its bound and unbound states (apo-state) are chosen to perform a total of 1.6 µs of molecular dynamics simulations. The simulated results show that the ECD of GLP-1R closes in the GLP-1 bound state and opens in the GLP-1 unbound state. To determine the critical role that GLP-1 played in regulating the open and closed states of the ECD, we applied the independent gradient model (IGM) to the simulation trajectories. We found that the "hand-like" N-terminal of the GLP-1R ECD plays an important role in the GLP-1 binding. In contrast, the apo-state GLP-1R ECD opens and exposes the two ligand binding domains of GLP-1 after 200 ns of simulations. To elucidate the open and closed mechanisms of GLP-1R ECD in the apo-state and GLP-1 bound state, the Markov state model (MSM) is performed on the MD simulation trajectories. Our results provide possible transition pathways from the closed state to open state of the apo-state GLP-1R ECD. Each pathway contains several intermediate states that correspond to different local minima in deep wells. The dynamical relationships and the most possible conversion pathway between two states are detailed through the MSM analysis. Our results profile the conformation transition mechanism of the GLP-1R ECD and will help in hypoglycemic peptide design of GLP-1R.