AlphaSpace 2.0: Representing Concave Biomolecular Surfaces Using β-Clusters

Diffusion Models in De Novo Drug Design

Diffusion models have emerged as powerful tools for molecular generation, particularly in the context of 3D molecular structures. Inspired by nonequilibrium statistical physics, these models can generate 3D molecular structures with specific properties or requirements crucial to drug discovery. Diffusion models were particularly successful at learning the complex probability distributions of 3D molecular geometries and their corresponding chemical and physical properties through forward and reverse diffusion processes. This review focuses on the technical implementation of diffusion models tailored for 3D molecular generation. It compares the performance, evaluation methods, and implementation details of various diffusion models used for molecular generation tasks. We cover strategies for atom and bond representation, architectures of reverse diffusion denoising networks, and challenges associated with generating stable 3D molecular structures. This review also explores the applications of diffusion models in de novo drug design and related areas of computational chemistry, such as structure-based drug design, including target-specific molecular generation, molecular docking, and molecular dynamics of protein-ligand complexes. We also cover conditional generation on physical properties, conformation generation, and fragment-based drug design. By summarizing the state-of-the-art diffusion models for 3D molecular generation, this review sheds light on their role in advancing drug discovery and their current limitations.

Subpocket-Based Analysis Approach for the Protein Pocket Dynamics

Structural and dynamic characteristics of protein pockets remarkably influence their biological functions and are also important for enzyme engineering and new drug research and development. To date, several softwares have been developed to analyze the dynamic properties of protein pockets. However, due to the complexity and diversity of the pocket information during the kinetic relaxation, further improvement and capacity expansion of current tools are required. Here, we developed a platform software AlphaTraj in which a computational strategy that divides the whole protein pocket into subpockets and examines various properties of the subpockets such as survival time, stability, and correlation was proposed and implemented. We also proposed a scoring function for the subpockets as well as the whole pocket to visualize the quality of the pocket. Furthermore, we implemented automated conformational search functions for ligand docking and ligand optimization. These functions may help us to gain a deep understanding of the dynamic properties of protein pockets and accelerate the protein engineering and the design of inhibitors and small-molecule drugs. The software is freely available at https://github.com/dooo12332/AlphaTraj.git under the GNU GPL license.

Integrated Molecular Modeling and Machine Learning for Drug Design

Modern therapeutic development often involves several stages that are interconnected, and multiple iterations are usually required to bring a new drug to the market. Computational approaches have increasingly become an indispensable part of helping reduce the time and cost of the research and development of new drugs. In this Perspective, we summarize our recent efforts on integrating molecular modeling and machine learning to develop computational tools for modulator design, including a pocket-guided rational design approach based on AlphaSpace to target protein-protein interactions, delta machine learning scoring functions for protein-ligand docking as well as virtual screening, and state-of-the-art deep learning models to predict calculated and experimental molecular properties based on molecular mechanics optimized geometries. Meanwhile, we discuss remaining challenges and promising directions for further development and use a retrospective example of FDA approved kinase inhibitor Erlotinib to demonstrate the use of these newly developed computational tools.

DataDTA: a multi-feature and dual-interaction aggregation framework for drug-target binding affinity prediction

MOTIVATION

Accurate prediction of drug-target binding affinity (DTA) is crucial for drug discovery. The increase in the publication of large-scale DTA datasets enables the development of various computational methods for DTA prediction. Numerous deep learning-based methods have been proposed to predict affinities, some of which only utilize original sequence information or complex structures, but the effective combination of various information and protein-binding pockets have not been fully mined. Therefore, a new method that integrates available key information is urgently needed to predict DTA and accelerate the drug discovery process.

RESULTS

In this study, we propose a novel deep learning-based predictor termed DataDTA to estimate the affinities of drug-target pairs. DataDTA utilizes descriptors of predicted pockets and sequences of proteins, as well as low-dimensional molecular features and SMILES strings of compounds as inputs. Specifically, the pockets were predicted from the three-dimensional structure of proteins and their descriptors were extracted as the partial input features for DTA prediction. The molecular representation of compounds based on algebraic graph features was collected to supplement the input information of targets. Furthermore, to ensure effective learning of multiscale interaction features, a dual-interaction aggregation neural network strategy was developed. DataDTA was compared with state-of-the-art methods on different datasets, and the results showed that DataDTA is a reliable prediction tool for affinities estimation. Specifically, the concordance index (CI) of DataDTA is 0.806 and the Pearson correlation coefficient (R) value is 0.814 on the test dataset, which is higher than other methods.

AVAILABILITY AND IMPLEMENTATION

The codes and datasets of DataDTA are available at https://github.com/YanZhu06/DataDTA.

Characterization of an allosteric inhibitor of fungal-specific C-24 sterol methyltransferase to treat Candida albicans infections

Filamentation is an important virulence factor of the pathogenic fungus Candida albicans. The abolition of Candida albicans hyphal formation by disrupting sterol synthesis is an important concept for the development of antifungal drugs with high safety. Here, we conduct a high-throughput screen using a C. albicans strain expressing green fluorescent protein-labeled Dpp3 to identify anti-hypha agents by interfering with ergosterol synthesis. The antipyrine derivative H55 is characterized to have minimal cytotoxicity and potent inhibition of C. albicans hyphal formation in multiple cultural conditions. H55 monotherapy exhibits therapeutic efficacy in mouse models of azole-resistant candidiasis. H55 treatment increases the accumulation of zymosterol, the substrate of C-24 sterol methyltransferase (Erg6). The results of enzyme assays, photoaffinity labeling, molecular simulation, mutagenesis, and cellular thermal shift assays support H55 as an allosteric inhibitor of Erg6. Collectively, H55, an inhibitor of the fungal-specific enzyme Erg6, holds potential to treat C. albicans infections.

Protein-Ligand Docking in the Machine-Learning Era

Molecular docking plays a significant role in early-stage drug discovery, from structure-based virtual screening (VS) to hit-to-lead optimization, and its capability and predictive power is critically dependent on the protein-ligand scoring function. In this review, we give a broad overview of recent scoring function development, as well as the docking-based applications in drug discovery. We outline the strategies and resources available for structure-based VS and discuss the assessment and development of classical and machine learning protein-ligand scoring functions. In particular, we highlight the recent progress of machine learning scoring function ranging from descriptor-based models to deep learning approaches. We also discuss the general workflow and docking protocols of structure-based VS, such as structure preparation, binding site detection, docking strategies, and post-docking filter/re-scoring, as well as a case study on the large-scale docking-based VS test on the LIT-PCBA data set.

Delta Machine Learning to Improve Scoring-Ranking-Screening Performances of Protein-Ligand Scoring Functions

Protein-ligand scoring functions are widely used in structure-based drug design for fast evaluation of protein-ligand interactions, and it is of strong interest to develop scoring functions with machine-learning approaches. In this work, by expanding the training set, developing physically meaningful features, employing our recently developed linear empirical scoring function Lin_F9 (Yang, C. J. Chem. Inf. Model. 2021, 61, 4630-4644) as the baseline, and applying extreme gradient boosting (XGBoost) with Δ-machine learning, we have further improved the robustness and applicability of machine-learning scoring functions. Besides the top performances for scoring-ranking-screening power tests of the CASF-2016 benchmark, the new scoring function Δ_{Lin_F9}XGB also achieves superior scoring and ranking performances in different structure types that mimic real docking applications. The scoring powers of Δ_{Lin_F9}XGB for locally optimized poses, flexible redocked poses, and ensemble docked poses of the CASF-2016 core set achieve Pearson's correlation coefficient (R) values of 0.853, 0.839, and 0.813, respectively. In addition, the large-scale docking-based virtual screening test on the LIT-PCBA data set demonstrates the reliability and robustness of Δ_{Lin_F9}XGB in virtual screening application. The Δ_{Lin_F9}XGB scoring function and its code are freely available on the web at (https://yzhang.hpc.nyu.edu/Delta_LinF9_XGB).

Identification of Secondary Binding Sites on Protein Surfaces for Rational Elaboration of Synthetic Protein Mimics

Minimal mimics of protein conformations provide rationally designed ligands to modulate protein function. The advantage of minimal mimics is that they can be chemically synthesized and coaxed to be proteolytically resistant; a key disadvantage is that minimization of the protein binding epitope may be associated with loss of affinity and specificity. Several approaches to overcome this challenge may be envisioned, including deployment of covalent warheads and use of nonnatural residues to improve contacts with the binding surface. Herein, we describe our computational and experimental efforts to enhance the minimal protein mimics with fragments that can contact undiscovered binding pockets on Mdm2 and MdmX-two well-studied protein partners of p53.

Computational methods-guided design of modulators targeting protein-protein interactions (PPIs)

Protein-protein interactions (PPIs) play a pivotal role in extensive biological processes and are thus crucial to human health and the development of disease states. Due to their critical implications, PPIs have been spotlighted as promising drug targets of broad-spectrum therapeutic interests. However, owing to the general properties of PPIs, such as flat surfaces, featureless conformations, difficult topologies, and shallow pockets, previous attempts were faced with serious obstacles when targeting PPIs and almost portrayed them as "intractable" for decades. To date, rapid progress in computational chemistry and structural biology methods has promoted the exploitation of PPIs in drug discovery. These techniques boost their cost-effective and high-throughput traits, and enable the study of dynamic PPI interfaces. Thus, computational methods represent an alternative strategy to target "undruggable" PPI interfaces and have attracted intense pharmaceutical interest in recent years, as exemplified by the accumulating number of successful cases. In this review, we first introduce a diverse set of computational methods used to design PPI modulators. Herein, we focus on the recent progress in computational strategies and provide a comprehensive overview covering various methodologies. Importantly, a list of recently-reported successful examples is highlighted to verify the feasibility of these computational approaches. Finally, we conclude the general role of computational methods in targeting PPIs, and also discuss future perspectives on the development of such aids.