Applications In Drug Design Research Articles

Machine learning approaches and their applications in drug discovery and design.

This review is focused on several machine learning approaches used in chemoinformatics. Machine learning approaches provide tools and algorithms to improve drug discovery. Many physicochemical properties of drugs like toxicity, absorption, drug-drug interaction, carcinogenesis, and distribution have been effectively modeled by QSAR techniques. Machine learning is a subset of artificial intelligence, and this technique has shown tremendous potential in the field of drug discovery. Techniques discussed in this review are capable of modeling non-linear datasets, as well as big data of increasing depth and complexity. Various machine learning-based approaches are being used for drug target prediction, modeling the structure of drug target, binding site prediction, ligand-based similarity searching, de novo designing of ligands with desired properties, developing scoring functions for molecular docking, building QSAR model for biological activity prediction, and prediction of pharmacokinetic and pharmacodynamic properties of ligands. In recent years, these predictive tools and models have achieved good accuracy. By the use of more related input data, relevant parameters, and appropriate algorithms, the accuracy of these predictions can be further improved.

A Consensus Compound/Bioactivity Dataset for Data-Driven Drug Design and Chemogenomics.

Publicly available compound and bioactivity databases provide an essential basis for data-driven applications in life-science research and drug design. By analyzing several bioactivity repositories, we discovered differences in compound and target coverage advocating the combined use of data from multiple sources. Using data from ChEMBL, PubChem, IUPHAR/BPS, BindingDB, and Probes & Drugs, we assembled a consensus dataset focusing on small molecules with bioactivity on human macromolecular targets. This allowed an improved coverage of compound space and targets, and an automated comparison and curation of structural and bioactivity data to reveal potentially erroneous entries and increase confidence. The consensus dataset comprised of more than 1.1 million compounds with over 10.9 million bioactivity data points with annotations on assay type and bioactivity confidence, providing a useful ensemble for computational applications in drug design and chemogenomics.

Recent Advances of Pyridinone in Medicinal Chemistry.

Pyridinones have been adopted as an important block in medicinal chemistry that could serve as hydrogen bond donors and acceptors. With the help of feasible synthesis routes via established condensation reactions, the physicochemical properties of such a scaffold could be manipulated by adjustment of polarity, lipophilicity, and hydrogen bonding, and eventually lead to its wide application in fragment-based drug design, biomolecular mimetics, and kinase hinge-binding motifs. In addition, most pyridinone derivatives exhibit various biological activities ranging from antitumor, antimicrobial, anti-inflammatory, and anticoagulant to cardiotonic effects. This review focuses on recent contributions of pyridinone cores to medicinal chemistry, and addresses the structural features and structure–activity relationships (SARs) of each drug-like molecule. These advancements contribute to an in-depth understanding of the potential of this biologically enriched scaffold and expedite the development of its new applications in drug discovery.

Out-of-the-box deep learning prediction of quantum-mechanical partial charges by graph representation and transfer learning.

Accurate prediction of atomic partial charges with high-level quantum mechanics (QM) methods suffers from high computational cost. Numerous feature-engineered machine learning (ML)-based predictors with favorable computability and reliability have been developed as alternatives. However, extensive expertise effort was needed for feature engineering of atom chemical environment, which may consequently introduce domain bias. In this study, SuperAtomicCharge, a data-driven deep graph learning framework, was proposed to predict three important types of partial charges (i.e. RESP, DDEC4 and DDEC78) derived from high-level QM calculations based on the structures of molecules. SuperAtomicCharge was designed to simultaneously exploit the 2D and 3D structural information of molecules, which was proved to be an effective way to improve the prediction accuracy of the model. Moreover, a simple transfer learning strategy and a multitask learning strategy based on self-supervised descriptors were also employed to further improve the prediction accuracy of the proposed model. Compared with the latest baselines, including one GNN-based predictor and two ML-based predictors, SuperAtomicCharge showed better performance on all the three external test sets and had better usability and portability. Furthermore, the QM partial charges of new molecules predicted by SuperAtomicCharge can be efficiently used in drug design applications such as structure-based virtual screening, where the predicted RESP and DDEC4 charges of new molecules showed more robust scoring and screening power than the commonly used partial charges. Finally, two tools including an online server (http://cadd.zju.edu.cn/deepchargepredictor) and the source code command lines (https://github.com/zjujdj/SuperAtomicCharge) were developed for the easy access of the SuperAtomicCharge services.

Research on Combining Pattern Mining and Evolutionary Algorithm for Critical Node Detection Problems

The critical node detection problems (CNDPs) have important applications in network security, smart grid, epidemic control, drug design, and risk assessment. The critical node problem (CNP) is one of well-known CNDPs, which is NP-hard. In this work, we combine frequent pattern mining with evolutionary algorithm for solving CNP, where pattern mined from high-quality solutions are used to guide the construction of offspring solution. More specifically, based on the memetic algorithm (Memetic Algorithm for CNP, MACNP), frequent pattern mining was integrated into the memetic algorithm framework. The frequent patterns mined were used to construct the offspring solution, instead of crossover operator in MACNP. In the research, the nodes in frequent items set were directly fixed as part of the offspring solution. So the offspring solution inherited excellent properties from more parent solutions, thus forming the algorithm called Frequent Pattern Based Search for CNP (FPBS-CNP). This strategy improved the algorithm efficiency. Experiments were conducted on synthetic and real-world benchmark. We experimentally compare FPBS-CNP with MACNP. The experiment results show that FPBS-CNP performs well on small instances and has potential in solving large instances.

In silico binding affinity prediction for metabotropic glutamate receptors using both endpoint free energy methods and a machine learning-based scoring function.

Metabotropic glutamate receptors (mGluRs) play an important role in regulating glutamate signal pathways, which are involved in neuropathy and periphery homeostasis. mGluR4, which belongs to Group III mGluRs, is most widely distributed in the periphery among all the mGluRs. It has been proved that the regulation of this receptor is involved in diabetes, colorectal carcinoma and many other diseases. However, the application of structure-based drug design to identify small molecules to regulate the mGluR4 receptor is limited due to the absence of a resolved mGluR4 protein structure. In this work, we first built a homology model of mGluR4 based on a crystal structure of mGluR8, and then conducted hierarchical virtual screening (HVS) to identify possible active ligands for mGluR4. The HVS protocol consists of three hierarchical filters including Glide docking, molecular dynamic (MD) simulation and binding free energy calculation. We successfully prioritized active ligands of mGluR4 from a set of screening compounds using HVS. The predicted active ligands based on binding affinities can almost cover all the experiment-determined active ligands, with only one ligand missed. The correlation between the measured and predicted binding affinities is significantly improved for the MM-PB/GBSA-WSAS methods compared to the Glide docking method. More importantly, we have identified hotspots for ligand binding, and we found that SER157 and GLY158 tend to contribute to the selectivity of mGluR4 ligands, while ALA154 and ALA155 could account for the ligand selectivity to mGluR8. We also recognized other 5 key residues that are critical for ligand potency. The difference of the binding profiles between mGluR4 and mGluR8 can guide us to develop more potent and selective modulators. Moreover, we evaluated the performance of IPSF, a novel type of scoring function trained by a machine learning algorithm on residue-ligand interaction profiles, in guiding drug lead optimization. The cross-validation root-mean-square errors (RMSEs) are much smaller than those by the endpoint methods, and the correlation coefficients are comparable to the best endpoint methods for both mGluRs. Thus, machine learning-based IPSF can be applied to guide lead optimization, albeit the total number of actives/inactives are not big, a typical scenario in drug discovery projects.

Application of Site-Identification by Ligand Competitive Saturation in Computer-Aided Drug Design.

Site Identification by Ligand Competitive Saturation (SILCS) is a molecular simulation approach that uses diverse small solutes in aqueous solution to obtain functional group affinity patterns of a protein or other macromolecule. This involves employing a combined Grand Canonical Monte Carlo (GCMC)-molecular dynamics (MD) method to sample the full 3D space of the protein, including deep binding pockets and interior cavities from which functional group free energy maps (FragMaps) are obtained. The information content in the maps, which include contributions from protein flexibilty and both protein and functional group desolvation contributions, can be used in many aspects of the drug discovery process. These include identification of novel ligand binding pockets, including allosteric sites, pharmacophore modeling, prediction of relative protein-ligand binding affinities for database screening and lead optimization efforts, evaluation of protein-protein interactions as well as in the formulation of biologics-based drugs including monoclonal antibodies. The present article summarizes the various tools developed in the context of the SILCS methodology and their utility in computer-aided drug design (CADD) applications, showing how the SILCS toolset can improve the drug-development process on a number of fronts with respect to both accuracy and throughput representing a new avenue of CADD applications.

Application of Machine Learning in Drug Discovery and Development Lifecycle

Machine learning and Artificial Intelligence have significantly advanced in recent years owing to their potential to considerably increase the quality of life while reducing human workload. The paper demonstrates how AI and ML are used in the drug development process to shorten and enhance the overall timeline. It contains pertinent information on a variety of Machine Learning approaches and algorithms that are used across the whole drug development process to speed up research, save expenses, and reduce risks related to clinical trials. A range of QSAR analysis, hit finding, and de novo drug design applications are used in the pharmaceutical industry to enhance decision-making. As technologies like high-throughput screening and computation analysis of databases used for lead and target identification and development create and integrate vast volumes of data, machine learning and deep learning have grown in importance. It has also been emphasized how these cognitive models and tools may be used in lead creation, optimization, and thorough virtual screening. In this paper, problem statements and the corresponding state-of-the-art models have been considered for target validation, prognostic biomarkers, and digital pathology. Machine Learning models play a vital role in the various operations related to clinical trials embracing protocol optimization, participant management, data analysis and storage, clinical trial data verification, and surveillance. Post-development drug monitoring and unique industrially prevalent ML applications of pharmacovigilance have also been discussed. As a result, the goal of this study is to investigate the machine learning and deep learning algorithms utilised across the drug development lifecycle as well as the supporting techniques that have the potential to be useful.

PyKVFinder: an efficient and integrable Python package for biomolecular cavity detection and characterization in data science

BackgroundBiomolecular interactions that modulate biological processes occur mainly in cavities throughout the surface of biomolecular structures. In the data science era, structural biology has benefited from the increasing availability of biostructural data due to advances in structural determination and computational methods. In this scenario, data-intensive cavity analysis demands efficient scripting routines built on easily manipulated data structures. To fulfill this need, we developed pyKVFinder, a Python package to detect and characterize cavities in biomolecular structures for data science and automated pipelines.ResultspyKVFinder efficiently detects cavities in biomolecular structures and computes their volume, area, depth and hydropathy, storing these cavity properties in NumPy arrays. Benefited from Python ecosystem interoperability and data structures, pyKVFinder can be integrated with third-party scientific packages and libraries for mathematical calculations, machine learning and 3D visualization in automated workflows. As proof of pyKVFinder’s capabilities, we successfully identified and compared ADRP substrate-binding site of SARS-CoV-2 and a set of homologous proteins with pyKVFinder, showing its integrability with data science packages such as matplotlib, NGL Viewer, SciPy and Jupyter notebook.ConclusionsWe introduce an efficient, highly versatile and easily integrable software for detecting and characterizing biomolecular cavities in data science applications and automated protocols. pyKVFinder facilitates biostructural data analysis with scripting routines in the Python ecosystem and can be building blocks for data science and drug design applications.

Tailored Modeling of Rivastigmine Derivatives as Dual Acetylcholinesterase and Butyrylcholinesterase Inhibitors for Alzheimer's Disease Treatment.

Rational modification of known drug candidates to design more potent ones using computational methods has found application in drug design, development, and discovery. Herein, we integrate computational and theoretical methodologies to unveil rivastigmine derivatives as dual inhibitors of acetylcholinesterase (AChE) and butyrylcholinesterase (BuChE) for Alzheimer's disease (AD) management. The investigation entails pharmacokinetics screening, density functional theory (DFT) mechanistic study, molecular docking, and molecular dynamics (MD) simulation. We designed over 20 rivastigmine substituents, subject them to some analyses, and identified RL2 with an appreciable blood-brain barrier score and no permeability glycoprotein binding. The compound shows higher acylation energy and a favored binding affinity to the cholinesterase enzymes. RL2 interacts with the AChE and BuChE active sites showing values of -41.1/-39.5 kcal mol-1 while rivastigmine binds with -32.7/-30.7 kcal mol-1 for these enzymes. The study revealed RL2 (4-fluorophenyl rivastigmine) as a potential dual inhibitor for AChE and BuChE towards Alzheimer's disorder management.

Bioisosteres of the Phenyl Ring: Recent Strategic Applications in Lead Optimization and Drug Design.

The benzene moiety is the most prevalent ring system in marketed drugs, underscoring its historic popularity in drug design either as a pharmacophore or as a scaffold that projects pharmacophoric elements. However, introspective analyses of medicinal chemistry practices at the beginning of the 21st century highlighted the indiscriminate deployment of phenyl rings as an important contributor to the poor physicochemical properties of advanced molecules, which limited their prospects of being developed into effective drugs. This Perspective deliberates on the design and applications of bioisosteric replacements for a phenyl ring that have provided practical solutions to a range of developability problems frequently encountered in lead optimization campaigns. While the effect of phenyl ring replacements on compound properties is contextual in nature, bioisosteric substitution can lead to enhanced potency, solubility, and metabolic stability while reducing lipophilicity, plasma protein binding, phospholipidosis potential, and inhibition of cytochrome P450 enzymes and the hERG channel.

Structure determination of GPCRs: cryo-EM compared with X-ray crystallography.

G protein-coupled receptors (GPCRs) are the largest single family of cell surface receptors encoded by the human genome and they play pivotal roles in co-ordinating cellular systems throughout the human body, making them ideal drug targets. Structural biology has played a key role in defining how receptors are activated and signal through G proteins and β-arrestins. The application of structure-based drug design (SBDD) is now yielding novel compounds targeting GPCRs. There is thus significant interest from both academia and the pharmaceutical industry in the structural biology of GPCRs as currently only about one quarter of human non-odorant receptors have had their structure determined. Initially, all the structures were determined by X-ray crystallography, but recent advances in electron cryo-microscopy (cryo-EM) now make GPCRs tractable targets for single-particle cryo-EM with comparable resolution to X-ray crystallography. So far this year, 78% of the 99 GPCR structures deposited in the PDB (Jan–Jul 2021) were determined by cryo-EM. Cryo-EM has also opened up new possibilities in GPCR structural biology, such as determining structures of GPCRs embedded in a lipid nanodisc and multiple GPCR conformations from a single preparation. However, X-ray crystallography still has a number of advantages, particularly in the speed of determining many structures of the same receptor bound to different ligands, an essential prerequisite for effective SBDD. We will discuss the relative merits of cryo-EM and X-ray crystallography for the structure determination of GPCRs and the future potential of both techniques.

Enabling Efficient Folding and High-Resolution Crystallographic Analysis of Bracelet Cyclotides.

Cyclotides have attracted great interest as drug design scaffolds because of their unique cyclic cystine knotted topology. They are classified into three subfamilies, among which the bracelet subfamily represents the majority and comprises the most bioactive cyclotides, but are the most poorly utilized in drug design applications. A long-standing challenge has been the very low in vitro folding yields of bracelets, hampering efforts to characterize their structures and activities. Herein, we report substantial increases in bracelet folding yields enabled by a single point mutation of residue Ile-11 to Leu or Gly. We applied this discovery to synthesize mirror image enantiomers and used quasi-racemic crystallography to elucidate the first crystal structures of bracelet cyclotides. This study provides a facile strategy to produce bracelet cyclotides, leading to a general method to easily access their atomic resolution structures and providing a basis for development of biotechnological applications.

Experimental and computational study of naphthalimide derivatives: Synthesis, optical, nonlinear optical and antiviral properties

The nonlinear optical (NLO) and antiviral properties of naphthalimide Schiff base compounds (5a–c) were experimentally and computationally investigated. The synthesized compounds (5a–c) were successfully characterized via UV–Vis, FTIR, 1H NMR, fluorescence spectroscopy, and elemental analysis. The calculated average third-order NLO polarizabilities (˂γ˃) of 5a, 5b, and 5c were found to be 5, 9, and 21 times greater than the ˂γ˃ amplitude of p-NA, respectively. The computed results revealed the potential of the synthesized compounds for NLO applications. Additionally, molecular docking studies of the synthesized compounds with two crucial SARS-CoV-2 proteins were performed to examine their biochemical properties. Compound 5c exhibited a higher binding affinity with the spike protein compared to that with Mᴾᴿᴼ. The results obtained herein indicate the potential of the synthesized naphthalimide derivatives for optoelectronic and drug design applications.

Accurate Estimation of Solvent Accessible Surface Area for Coarse-Grained Biomolecular Structures with Deep Learning.

Coarse-grained (CG) models of biomolecules have been widely used in protein/ribonucleic acid (RNA) three-dimensional structure prediction, docking, drug design, and molecular simulations due to their superiority in computational efficiency. Most of these applications strongly depend on the reasonable estimation of solvation free energy, which requires the accurate calculation of solvent accessible surface area (SASA). Although algorithms for SASA calculations with all-atom protein and RNA structures have been well-established, accurately estimating the SASA based on CG structures is extremely challenging. In this work, we developed a deep learning-based SASA estimator (DeepCGSA), which can provide almost perfect SASA estimation based on CG structures of protein and RNA molecules. Extensive testing analysis showed that for three types of widely used CG protein models, including the Cα-based, Cα-Cβ, and Martini models, the correlation coefficients between the predicted values and the reference values can be as high as 0.95-0.99, which perform dramatically better than available methods. In addition, the new method can be used for CG RNA structures and unfolded protein structures with much improved accuracy. We anticipate that DeepCGSA will be highly useful in the protein/RNA structure prediction, drug design, and other applications, in which accurate estimations of SASA for CG biomolecular structures are critically important.

Synthesis of Enantioenriched gem-Disubstituted 4-Imidazolidinones by Palladium-Catalyzed Decarboxylative Asymmetric Allylic Alkylation.

A variety of enantioenriched gem-disubstituted 4-imidazolidinones were prepared in up to >99% yield and 95% ee by the Pd-catalyzed decarboxylative asymmetric allylic alkylation of imidazolidinone-derived β-amidoesters. In the process of preparing these substrates, a rapid synthetic route to 4-imidazolidinone derivatives was developed, beginning from 2-thiohydantoin. The orthogonality of the benzoyl imide and tert-butyl carbamate groups used to protect these nitrogen-rich products was demonstrated, enabling potential applications in drug design.

DeepChargePredictor: a web server for predicting QM-based atomic charges via state-of-the-art machine-learning algorithms.

High-level quantum mechanics (QM) methods are no doubt the most reliable approaches for the prediction of atomic charges, but it usually needs very large computational resources, which apparently hinders the use of high-quality atomic charges in large-scale molecular modeling, such as high-throughput virtual screening. To solve this problem, several algorithms based on machine-learning (ML) have been developed to fit high-level QM atomic charges. Here, we proposed DeepChargePredictor, a web server that is able to generate the high-level QM atomic charges for small molecules based on two state-of-the-art ML algorithms developed in our group, namely AtomPathDescriptor and DeepAtomicCharge. These two algorithms were seamlessly integrated into the platform with the capability to predict three kinds of charges (i.e. RESP, AM1-BCC and DDEC) widely used in structure-based drug design. Moreover, we have comprehensively evaluated the performance of these charges generated by DeepChargePredictor for large-scale drug design applications, such as end-point binding free energy calculations and virtual screening, which all show reliable or even better performance compared with the baseline methods. The data in the article can be obtained on the web page http://cadd.zju.edu.cn/deepchargepredictor/publication. Supplementary data are available at Bioinformatics online.

1,3-Diazepine: A privileged scaffold in medicinal chemistry.

Privileged structures have been widely used as effective templates for drug discovery. While benzo-1,4-diazepine constitutes the first historical example of such a structure, the 1,3 analogue is just as rich in terms of applications in medicinal chemistry. The 1,3-diazepine moiety is present in numerous biological active compounds including natural products, and is used to design compounds displaying a large range of biological activities. It is present in the clinically used anticancer compound pentostatin, in several recent FDA approved β-lactamase inhibitors (e.g., avibactam) and also in coformycin, a natural product known as a ring-expanded purine analogue displaying antiviral and anticancer activities. Several other 1,3-diazepine containing compounds have entered into clinical trials. This heterocyclic structure has been and is still widely used in medicinal chemistry to design enzyme inhibitors, GPCR ligands, and so forth. This review endeavours to highlight the main use of the 1,3-diazepine scaffold and its derivatives, and their applications in medicinal chemistry, drug design, and therapy. We will focus more particularly on the development of enzyme inhibitors incorporating this scaffold, with a strong emphasis on the molecular interactions involved in the inhibition mechanism.

Gaussian accelerated molecular dynamics (GaMD): principles and applications.

Gaussian accelerated molecular dynamics (GaMD) is a robust computational method for simultaneous unconstrained enhanced sampling and free energy calculations of biomolecules. It works by adding a harmonic boost potential to smooth biomolecular potential energy surface and reduce energy barriers. GaMD greatly accelerates biomolecular simulations by orders of magnitude. Without the need to set predefined reaction coordinates or collective variables, GaMD provides unconstrained enhanced sampling and is advantageous for simulating complex biological processes. The GaMD boost potential exhibits a Gaussian distribution, thereby allowing for energetic reweighting via cumulant expansion to the second order (i.e., "Gaussian approximation"). This leads to accurate reconstruction of free energy landscapes of biomolecules. Hybrid schemes with other enhanced sampling methods, such as the replica exchange GaMD (rex-GaMD) and replica exchange umbrella sampling GaMD (GaREUS), have also been introduced, further improving sampling and free energy calculations. Recently, new "selective GaMD" algorithms including the ligand GaMD (LiGaMD) and peptide GaMD (Pep-GaMD) enabled microsecond simulations to capture repetitive dissociation and binding of small-molecule ligands and highly flexible peptides. The simulations then allowed highly efficient quantitative characterization of the ligand/peptide binding thermodynamics and kinetics. Taken together, GaMD and its innovative variants are applicable to simulate a wide variety of biomolecular dynamics, including protein folding, conformational changes and allostery, ligand binding, peptide binding, protein-protein/nucleic acid/carbohydrate interactions, and carbohydrate/nucleic acid interactions. In this review, we present principles of the GaMD algorithms and recent applications in biomolecular simulations and drug design.

Mapping the Epitope Interactions of Trastuzumab Via Plasma Induced Modification of Biomolecules (PLIMB)

Epitope mapping is the process of identifying the site(s) of an antigen where an antibody binds. Knowledge of epitopes facilitates drug design and patent applications; hence it is a key element of vaccine and drug development. Hydroxyl radical protein footprinting coupled with mass spectrometry offers opportunities to map antibody-binding regions of antigens with greater simplicity and increased throughput over x-ray crystallography. We report here the use of hydroxyl radical protein footprinting via Plasma Induced Modification of Biomolecules (PLIMB) to characterize the epitopes of trastuzumab (Herceptin) bound to receptor tyrosine-protein kinase erbB-2 (HER2). The protein samples were exposed to plasma for hydroxyl radical labeling, the labeled sites were identified using mass spectrometry, and the data was analyzed using Byos (Protein Metrics). The PLIMB data is compared to previously published results that determined the epitope of trastuzumab using x-ray crystallography.