Improving protein-ligand docking and screening accuracies by incorporating a scoring function correction term

High-confidence structural annotation of substances via multi-layer molecular network reveals the system-wide constituent alternations in milk interfered with diphenylolpropane

The spectral database-based mass spectrometry (MS) matching strategy is versatile for structural annotating in ingredient fluctuation profiling mediated by external interferences. However, the systematic variability of MS pool attributable to aliasing peaks and inadequacy of present spectral database resulted in a substantial metabolic feature depletion. An amended procedure termed multiple-charges overlap peaks extraction algorithm (MCOP) was proposed involving identifying collision-trigged dissociation precursor ions through iteratively matching mass features of fragmentations to expand the spectral reference library. We showcased the versatility and utility of established strategy in an investigation centered on the stimulation of milk mediated by diphenylolpropane (BPA). MCOP enabled efficient unknown annotations at metabolite-lipid-protein level, which elevated the accuracy of substance annotation to 85.3% after manual validation. Arginase and α-amylase (|r| > 0.75, p < 0.05) were first identified as the crucial issues via graph neural network-based virtual screening in the abnormal metabolism of urea triggered by BPA, resulting in the accumulation of arginine (original: 1.7 μg kg-1 1.7 times) and maltodextrin (original: 6.9 μg kg-1 2.9 times) and thus, exciting the potential dietary risks. Conclusively, MCOP demonstrated generalisation and scalability and substantially advanced the discovery of unknown metabolites for complex matrix samples, thus deciphering dark matter in multi-omics.

Water Network-Augmented Two-State Model for Protein-Ligand Binding Affinity Prediction

Water network rearrangement from the ligand-unbound state to the ligand-bound state is known to have significant effects on the protein-ligand binding interactions, but most of the current machine learning-based scoring functions overlook these effects. In this study, we endeavor to construct a comprehensive and realistic deep learning model by incorporating water network information into both ligand-unbound and -bound states. In particular, extended connectivity interaction features were integrated into graph representation, and graph transformer operator was employed to extract features of the ligand-unbound and -bound states. Through these efforts, we developed a water network-augmented two-state model called ECIFGraph::HM-Holo-Apo. Our new model exhibits satisfactory performance in terms of scoring, ranking, docking, screening, and reverse screening power tests on the CASF-2016 benchmark. In addition, it can achieve superior performance in large-scale docking-based virtual screening tests on the DEKOIS2.0 data set. Our study highlights that the use of a water network-augmented two-state model can be an effective strategy to bolster the robustness and applicability of machine learning-based scoring functions, particularly for targets with hydrophilic or solvent-exposed binding pockets.

An ensemble-based approach to estimate confidence of predicted protein-ligand binding affinity values

When designing a machine learning-based scoring function, we access a limited number of protein-ligand complexes with experimentally determined binding affinity values, representing only a fraction of all possible protein-ligand complexes. Consequently, it is crucial to report a measure of confidence and quantify the uncertainty in the model's predictions during test time. Here, we adopt the conformal prediction technique to evaluate the confidence of a prediction for each member of the core set of the CASF 2016 benchmark. The conformal prediction technique requires a diverse ensemble of predictors for uncertainty estimation. To this end, we introduce ENS-Score as an ensemble predictor, which includes 30 models with different protein-ligand representation approaches and achieves Pearson's correlation of 0.842 on the core set of the CASF 2016 benchmark. Also, we comprehensively investigate the residual error of each data point to assess the normality behavior of the distribution of the residual errors and their correlation to the structural features of the ligands, such as hydrophobic interactions and halogen bonding. In the end, we provide a local host web application to facilitate the usage of ENS-Score. All codes to repeat results are provided at https://github.com/miladrayka/ENS_Score.

A new paradigm for applying deep learning to protein-ligand interaction prediction

Protein-ligand interaction prediction presents a significant challenge in drug design. Numerous machine learning and deep learning (DL) models have been developed to accurately identify docking poses of ligands and active compounds against specific targets. However, current models often suffer from inadequate accuracy or lack practical physical significance in their scoring systems. In this research paper, we introduce IGModel, a novel approach that utilizes the geometric information of protein-ligand complexes as input for predicting the root mean square deviation of docking poses and the binding strength (pKd, the negative value of the logarithm of binding affinity) within the same prediction framework. This ensures that the output scores carry intuitive meaning. We extensively evaluate the performance of IGModel on various docking power test sets, including the CASF-2016 benchmark, PDBbind-CrossDocked-Core and DISCO set, consistently achieving state-of-the-art accuracies. Furthermore, we assess IGModel's generalizability and robustness by evaluating it on unbiased test sets and sets containing target structures generated by AlphaFold2. The exceptional performance of IGModel on these sets demonstrates its efficacy. Additionally, we visualize the latent space of protein-ligand interactions encoded by IGModel and conduct interpretability analysis, providing valuable insights. This study presents a novel framework for DL-based prediction of protein-ligand interactions, contributing to the advancement of this field. The IGModel is available at GitHub repository https://github.com/zchwang/IGModel.

RmsdXNA: RMSD prediction of nucleic acid-ligand docking poses using machine-learning method

Small molecule drugs can be used to target nucleic acids (NA) to regulate biological processes. Computational modeling methods, such as molecular docking or scoring functions, are commonly employed to facilitate drug design. However, the accuracy of the scoring function in predicting the closest-to-native docking pose is often suboptimal. To overcome this problem, a machine learning model, RmsdXNA, was developed to predict the root-mean-square-deviation (RMSD) of ligand docking poses in NA complexes. The versatility of RmsdXNA has been demonstrated by its successful application to various complexes involving different types of NA receptors and ligands, including metal complexes and short peptides. The predicted RMSD by RmsdXNA was strongly correlated with the actual RMSD of the docked poses. RmsdXNA also outperformed the rDock scoring function in ranking and identifying closest-to-native docking poses across different structural groups and on the testing dataset. Using experimental validated results conducted on polyadenylated nuclear element for nuclear expression triplex, RmsdXNA demonstrated better screening power for the RNA-small molecule complex compared to rDock. Molecular dynamics simulations were subsequently employed to validate the binding of top-scoring ligand candidates selected by RmsdXNA and rDock on MALAT1. The results showed that RmsdXNA has a higher success rate in identifying promising ligands that can bind well to the receptor. The development of an accurate docking score for a NA-ligand complex can aid in drug discovery and development advancements. The code to use RmsdXNA is available at the GitHub repository https://github.com/laiheng001/RmsdXNA.

Effect of Cu(II) compound containing dipicolinic acid on DNA damage: a study of antiproliferative activity and DNA interaction properties by spectroscopic, molecular docking and molecular dynamics approaches

A polymeric compound formulized as [Cu(µ-dipic)2{Na2(µ-H2O)4]n.2nH2O (I), where dipic is 2,6-pyridine dicarboxylic acid (dipicolinic acid, H2dipic), was synthesized by sonochemical irradiation. The initial in-vitro cytotoxic activity of this complex compared with renowned anticancer drugs like cisplatin, versus HCT116 colon cell lines, shows promising results. This study investigated the interaction mode between compound (I) and calf-thymus DNA utilizing a range of analytical techniques including spectrophotometry, fluorimetry, partition coefficient analysis, viscometry, gel electrophoresis and molecular docking technique. The results obtained from experimental methods reveal complex (I) could bind to CT-DNA via hydrogen bonding and van der Waals forces and the theoretical methods support it. Also, complex (I) indicates nuclease activity in the attendance of H2O2 and can act as an artificial nuclease to cleave DNA with high efficiency.Communicated by Ramaswamy H. Sarma.

Iso-mukaadial acetate and ursolic acid acetate bind to Plasmodium Falciparum heat shock protein 70: towards targeting parasite protein folding pathway

Plasmodium falciparum is the most lethal malaria parasite. P. falciparum Hsp70 (PfHsp70) is an essential molecular chaperone (facilitates protein folding) and is deemed a prospective antimalarial drug target. The present study investigates the binding capabilities of select plant derivatives, iso-mukaadial acetate (IMA) and ursolic acid acetate (UAA), against P. falciparum using an in silico docking approach. The interaction between the ligands and PfHsp70 was evaluated using plasmon resonance (SPR) analysis. Molecular docking, binding free energy analysis and molecular dynamics simulations were conducted towards understanding the mechanisms by which the compounds bind to PfHsp70. The molecular docking results revealed ligand flexibilities, conformations and positions of key amino acid residues and protein-ligand interactions as crucial factors accounting for selective inhibition of Hsp70. The simulation results also suggest protein-ligand van der Waals forces as the driving force guiding the interaction of these compounds with PfHsp70. Of the two compounds, UAA and IMA bound to PfHsp70 within the micromolar range based on surface plasmon resonance (SPR) based binding assay. Our findings pave way for future rational design of new selective compounds targeting PfHsp70.

TopoFormer: Multiscale Topology-enabled Structure-to-Sequence Transformer for Protein-Ligand Interaction Predictions

Pre-trained deep Transformers have had tremendous success in a wide variety of disciplines. However, in computational biology, essentially all Transformers are built upon the biological sequences, which ignores vital stereochemical information and may result in crucial errors in downstream predictions. On the other hand, three-dimensional (3D) molecular structures are incompatible with the sequential architecture of Transformer and natural language processing (NLP) models in general. This work addresses this foundational challenge by a topological Transformer (TopoFormer). TopoFormer is built by integrating NLP and a multiscale topology techniques, the persistent topological hyperdigraph Laplacian (PTHL), which systematically converts intricate 3D protein-ligand complexes at various spatial scales into a NLP-admissible sequence of topological invariants and homotopic shapes. Element-specific PTHLs are further developed to embed crucial physical, chemical, and biological interactions into topological sequences. TopoFormer surges ahead of conventional algorithms and recent deep learning variants and gives rise to exemplary scoring accuracy and superior performance in ranking, docking, and screening tasks in a number of benchmark datasets. The proposed topological sequences can be extracted from all kinds of structural data in data science to facilitate various NLP models, heralding a new era in AI-driven discovery.

CarsiDock: a deep learning paradigm for accurate protein-ligand docking and screening based on large-scale pre-training

The expertise accumulated in deep neural network-based structure prediction has been widely transferred to the field of protein-ligand binding pose prediction, thus leading to the emergence of a variety of deep learning-guided docking models for predicting protein-ligand binding poses without relying on heavy sampling. However, their prediction accuracy and applicability are still far from satisfactory, partially due to the lack of protein-ligand binding complex data. To this end, we create a large-scale complex dataset containing ∼9 M protein-ligand docking complexes for pre-training, and propose CarsiDock, the first deep learning-guided docking approach that leverages pre-training of millions of predicted protein-ligand complexes. CarsiDock contains two main stages, i.e., a deep learning model for the prediction of protein-ligand atomic distance matrices, and a translation, rotation and torsion-guided geometry optimization procedure to reconstruct the matrices into a credible binding pose. The pre-training and multiple innovative architectural designs facilitate the dramatically improved docking accuracy of our approach over the baselines in terms of multiple docking scenarios, thereby contributing to its outstanding early recognition performance in several retrospective virtual screening campaigns. Further explorations demonstrate that CarsiDock can not only guarantee the topological reliability of the binding poses but also successfully reproduce the crucial interactions in crystalized structures, highlighting its superior applicability.

Skin hydration modulatory activities of Ficus deltoidea extract

Ficus deltoidea was known for its potent antioxidant, anti-melanogenic and photoprotective skin barrier activities. These properties are contributed by its biomarkers which are vitexin and isovitexin. This study aims to optimize the yield of methanolic extraction of Ficus deltoidea leaves (EFD) and evaluate their effects on skin barrier function and hydration. For optimization, Box-Behnken design was utilized to investigate the effects of methanol concentration, sonication time, and solvent-to-sample ratio on the yields of vitexin and isovitexin in EFD. The optimal yields obtained were 32.29 mg/g for vitexin and 35.87 mg/g for isovitexin. The optimum extraction conditions were 77.66% methanol concentration, 20.03 min sonication time, and 19.88 mL/g solvent-to-sample ratio. The quantitative real-time polymerase chain reaction was utilized to measure variant marker genes of transglutaminase-1, caspase 14, ceramide synthase 3, involucrin, and filaggrin of EFD-induced keratinocyte differentiation by in vitro study. Exposure to EFD has elevated the mRNA levels of all tested marker genes by 0.7-9.2 folds. Then, in vivo efficacy study was conducted on 20 female subjects for 14 days to evaluate skin biophysical assessment of hydration. EFD topical formulation treatment successfully increased skin hydration on day 7 (43.74%) and day 14 (47.23%). In silico study by molecular docking was performed to identify intermolecular binding interactions of vitexin and isovitexin with the interested proteins of tested marker genes. The result of molecular docking to the interested proteins revealed a similar trend with real-time PCR data. In conclusion, EFD potentially enhanced the skin barrier function and hydration of human skin cells.

zPoseScore model for accurate and robust protein-ligand docking pose scoring in CASP15

We introduce a deep learning-based ligand pose scoring model called zPoseScore for predicting protein-ligand complexes in the 15th Critical Assessment of Protein Structure Prediction (CASP15). Our contributions are threefold: first, we generate six training and evaluation data sets by employing advanced data augmentation and sampling methods. Second, we redesign the "zFormer" module, inspired by AlphaFold2's Evoformer, to efficiently describe protein-ligand interactions. This module enables the extraction of protein-ligand paired features that lead to accurate predictions. Finally, we develop the zPoseScore framework with zFormer for scoring and ranking ligand poses, allowing for atomic-level protein-ligand feature encoding and fusion to output refined ligand poses and ligand per-atom deviations. Our results demonstrate excellent performance on various testing data sets, achieving Pearson's correlation R  = 0.783 and 0.659 for ranking docking decoys generated based on experimental and predicted protein structures of CASF-2016 protein-ligand complexes. Additionally, we obtain an averaged local distance difference test (lDDT pli = 0.558) of AIchemy LIG2 in CASP15 for de novo protein-ligand complex structure predictions. Detailed analysis shows that accurate ligand binding site prediction and side-chain orientation are crucial for achieving better prediction performance. Our proposed model is one of the most accurate protein-ligand pose prediction models and could serve as a valuable tool in small molecule drug discovery.

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.

Synthesis of Pyrrolo[3,4-b]pyridin-5-ones via Ugi-Zhu Reaction and In Vitro-In Silico Studies against Breast Carcinoma

An Ugi-Zhu three-component reaction (UZ-3CR) coupled in a one-pot manner to a cascade process (N-acylation/aza Diels-Alder cycloaddition/decarboxylation/dehydration) was performed to synthesize a series of pyrrolo[3,4-b]pyridin-5-ones in 20% to 92% overall yields using ytterbium triflate as a catalyst, toluene as a solvent, and microwaves as a heat source. The synthesized molecules were evaluated in vitro against breast cancer cell lines MDA-MB-231 and MCF-7, finding that compound 1f, at a concentration of 6.25 μM, exhibited a potential cytotoxic effect. Then, to understand the interactions between synthesized compounds and the main proteins related to the cancer cell lines, docking studies were performed on the serine/threonine kinase 1 (AKT1) and Orexetine type 2 receptor (Ox2R), finding moderate to strong binding energies, which matched accurately with the in vitro results. Additionally, molecular dynamics were performed between proteins related to the studied cell lines and the three best ligands.

Identification of potential flavonoid compounds as antibacterial therapeutics against Klebsiella pneumoniae infection using structure-based virtual screening and molecular dynamics simulation

Klebsiella pneumoniae, which is among the top three pathogens on WHO's priority list, is one of the gram-negative bacteria that doctors and researchers around the world have fought for decades. Capsular polysaccharide (CPS) protein is extensively recognized as an important K. pneumoniae virulence factor. Thus, CPS has become the most characterized target for the discovery of novel drug candidates. The ineffectiveness of currently existing antibiotics urges the search for potent antimicrobial compounds. Flavonoids are a group of plant metabolites that have antibacterial potential and can enhance the present medications to elicit improved results against diverse diseases without adverse reactions. Henceforth, the present study aims to illustrate the inhibitory potential of flavonoids with varying pharmacological properties, targeting the CPS protein of K. pneumoniae by in silico approaches. The flavonoid compounds (n = 169) were retrieved from the PubChem database and screened using the structure-based virtual screening approach. Compounds with the highest binding score were estimated through their pharmacokinetic effects by ADMET descriptors. Finally, four potential inhibitors with PubChem CID: (4301534, 5213, 5481948, and 637080) were selected after molecular docking and drug-likeness analysis. All four lead compounds were employed for the MDS analysis of a 100 ns time period. Various studies were undertaken to assess the stability of the protein-ligand complexes. The binding free energy was computed using MM-PBSA, and the outcomes indicated that the molecules are having stable interactions with the binding site of the target protein. The results revealed that all four compounds can be employed as potential therapeutics against K. pneumoniae.

Ligand binding affinity prediction with fusion of graph neural networks and 3D structure-based complex graph

Accurate prediction of protein-ligand binding affinity is pivotal for drug design and discovery. Here, we proposed a novel deep fusion graph neural networks framework named FGNN to learn the protein-ligand interactions from the 3D structures of protein-ligand complexes. Unlike 1D sequences for proteins or 2D graphs for ligands, the 3D graph of protein-ligand complex enables the more accurate representations of the protein-ligand interactions. Benchmark studies have shown that our fusion models FGNN can achieve more accurate prediction of binding affinity than any individual algorithm. The advantages of fusion strategies have been demonstrated in terms of expressive power of data, learning efficiency and model interpretability. Our fusion models show satisfactory performances on diverse data sets, demonstrating their generalization ability. Given the good performances in both binding affinity prediction and virtual screening, our fusion models are expected to be practically applied for drug screening and design. Our work highlights the potential of the fusion graph neural network algorithm in solving complex prediction problems in computational biology and chemistry. The fusion graph neural networks (FGNN) model is freely available in https://github.com/LinaDongXMU/FGNN.

In Silico Neuroprotective Effects of Specific Rheum palmatum Metabolites on Parkinson's Disease Targets

Parkinson's disease (PD) is one of the large-scale health issues detrimental to human quality of life, and current treatments are only focused on neuroprotection and easing symptoms. This study evaluated in silico binding activity and estimated the stability of major metabolites in the roots of R. palmatum (RP) with main protein targets in Parkinson's disease and their ADMET properties. The major metabolites of RP were subjected to molecular docking and QSAR with α-synuclein, monoamine oxidase isoform B, catechol o-methyltransferase, and A2A adenosine receptor. From this, emodin had the greatest binding activity with Parkinson's disease targets. The chemical stability of the selected compounds was estimated using density functional theory analyses. The docked compounds showed good stability for inhibitory action compared to dopamine and levodopa. According to their structure-activity relationship, aloe-emodin, chrysophanol, emodin, and rhein exhibited good inhibitory activity to specific targets. Finally, mediocre pharmacokinetic properties were observed due to unexceptional blood-brain barrier penetration and safety profile. It was revealed that the major metabolites of RP may have good neuroprotective activity as an additional hit for PD drug development. Also, an association between redox-mediating and activities with PD-relevant protein targets was observed, potentially opening discussion on electrochemical mechanisms with biological functions.

A generalized protein-ligand scoring framework with balanced scoring, docking, ranking and screening powers

Applying machine learning algorithms to protein-ligand scoring functions has aroused widespread attention in recent years due to the high predictive accuracy and affordable computational cost. Nevertheless, most machine learning-based scoring functions are only applicable to a specific task, e.g., binding affinity prediction, binding pose prediction or virtual screening, suggesting that the development of a scoring function with balanced performance in all critical tasks remains a grand challenge. To this end, we propose a novel parameterization strategy by introducing an adjustable binding affinity term that represents the correlation between the predicted outcomes and experimental data into the training of mixture density network. The resulting residue-atom distance likelihood potential not only retains the superior docking and screening power over all the other state-of-the-art approaches, but also achieves a remarkable improvement in scoring and ranking performance. We emphatically explore the impacts of several key elements on prediction accuracy as well as the task preference, and demonstrate that the performance of scoring/ranking and docking/screening tasks of a certain model could be well balanced through an appropriate manner. Overall, our study highlights the potential utility of our innovative parameterization strategy as well as the resulting scoring framework in future structure-based drug design.

Molecular Modeling Unveils the Effective Interaction of B-RAF Inhibitors with Rare B-RAF Insertion Variants

The Food and Drug Administration (FDA) has approved MAPK inhibitors as a treatment for melanoma patients carrying a mutation in codon V600 of the BRAF gene exclusively. However, BRAF mutations outside the V600 codon may occur in a small percentage of melanomas. Although these rare variants may cause B-RAF activation, their predictive response to B-RAF inhibitor treatments is still poorly understood. We exploited an integrated approach for mutation detection, tumor evolution tracking, and assessment of response to treatment in a metastatic melanoma patient carrying the rare p.T599dup B-RAF mutation. He was addressed to Dabrafenib/Trametinib targeted therapy, showing an initial dramatic response. In parallel, in-silico ligand-based homology modeling was set up and performed on this and an additional B-RAF rare variant (p.A598_T599insV) to unveil and justify the success of the B-RAF inhibitory activity of Dabrafenib, showing that it could adeptly bind both these variants in a similar manner to how it binds and inhibits the V600E mutant. These findings open up the possibility of broadening the spectrum of BRAF inhibitor-sensitive mutations beyond mutations at codon V600, suggesting that B-RAF V600 WT melanomas should undergo more specific investigations before ruling out the possibility of targeted therapy.

In silico evaluation of binding interaction and ADME study of new 1,3-diazetidin-2-one derivatives with high antiproliferative activity

A series of eight novels' 1,3-diazetidin-2-ones have been proposed to assess their potential activities. They are intended to examine antiproliferative effects through inhibition of epidermal growth factor receptor (EGFR) expression. These eight compounds strongly interact with the EGFR protein, responsible for the activity. As part of a present study, these compounds were docked to the crystal structure of the EGFR (Protein Data Bank code: 1 M17) to determine their binding affinity at the active site. Based on computer predictions, two compounds were demonstrated high scores of 80.80 and 85.89. After analyzing ADME properties, these compounds were found to have significant potential for binding. Consequently, the abilities of gefitinib, erlotinib, imatinib, and sorafenib were selected for comparison as controls. Computational methods were performed to predict the critical disposition of eight novels' 1,3-diazetidin-2-one derivatives to the EGFR. Moreover, a docking technique employing the Genetic Optimization for Ligand Docking program was conducted. Compounds 2 and 7 demonstrate a high docking peace-wise scoring function (PLP) fitness of 85.89 and 80.80, respectively. They fulfilled the Lipinski's rule, topological descriptors, and fingerprints of drug-like molecular structure keys. These compounds can be used as lead compounds to develop novel antiproliferative agents. The outcome of applying this study is novel series of 1,3-diazetidin-2-one compounds as new analogs were designed and evaluated for their antiproliferative activity with a higher potency profile and binding affinity within the active sites of EGFR.

Conserved Sites and Recognition Mechanisms of T1R1 and T2R14 Receptors Revealed by Ensemble Docking and Molecular Descriptors and Fingerprints Combined with Machine Learning

Taste peptides, as an important component of protein-rich foodstuffs, potentiate the nutrition and taste of food. Thereinto, umami- and bitter-taste peptides have been ex tensively reported, while their taste mechanisms remain unclear. Meanwhile, the identification of taste peptides is still a time-consuming and costly task. In this study, 489 peptides with umami/bitter taste from TPDB (http://tastepeptides-meta.com/) were collected and used to train the classification models based on docking analysis, molecular descriptors (MDs), and molecular fingerprints (FPs). A consensus model, taste peptide docking machine (TPDM), was generated based on five learning algorithms (linear regression, random forest, gaussian naive bayes, gradient boosting tree, and stochastic gradient descent) and four molecular representation schemes. Model interpretive analysis showed that MDs (VSA_EState, MinEstateIndex, MolLogP) and FPs (598, 322, 952) had the greatest impact on the umami/bitter prediction of peptides. Based on the consensus docking results, we obtained the key recognition modes of umami/bitter receptors (T1Rs/T2Rs): (1) residues 107S-109S, 148S-154T, 247F-249A mainly form hydrogen bonding contacts and (2) residues 153A-158L, 163L, 181Q, 218D, 247F-249A in T1R1 and 56D, 106P, 107V, 152V-156F, 173K-180F in T2R14 constituted their hydrogen bond pockets. The model is available at http://www.tastepeptides-meta.com/yyds.

Geometric Interaction Graph Neural Network for Predicting Protein-Ligand Binding Affinities from 3D Structures (GIGN)

Predicting protein-ligand binding affinities (PLAs) is a core problem in drug discovery. Recent advances have shown great potential in applying machine learning (ML) for PLA prediction. However, most of them omit the 3D structures of complexes and physical interactions between proteins and ligands, which are considered essential to understanding the binding mechanism. This paper proposes a geometric interaction graph neural network (GIGN) that incorporates 3D structures and physical interactions for predicting protein-ligand binding affinities. Specifically, we design a heterogeneous interaction layer that unifies covalent and noncovalent interactions into the message passing phase to learn node representations more effectively. The heterogeneous interaction layer also follows fundamental biological laws, including invariance to translations and rotations of the complexes, thus avoiding expensive data augmentation strategies. GIGN achieves state-of-the-art performance on three external test sets. Moreover, by visualizing learned representations of protein-ligand complexes, we show that the predictions of GIGN are biologically meaningful.

Structure-based virtual screening for potent inhibitors of GH-20 β-N-acetylglucosaminidase: classical and machine learning scoring functions, and molecular dynamics simulations

GH-20 β-N-acetylglucosaminidases (GlcNAcases) are promising targets in the development of antimicrobial agents against Vibrio infections in humans and aquatic animals. In this study, we set up structure-based virtual screening to identify potential GH-20 GlcNAcase inhibitors from the Reaxys commercial database, using VhGlcNAcase from V. campbellii type strain ATCC® BAA 1116 as the protein target and Redoxal as the reference ligand. Using ChemPLP and RF-Score-VS machine learning scoring functions, eight lead compounds were identified and further evaluated for protein interaction preference and pharmacological properties. Protein-ligand analysis demonstrated that all selected compounds interacted exclusively at subsite - 1 with five hydrophobic residues W487, W505, W546, W582 and V544 at site S1, and with two polar residues, D437 and E438, at site 3. For subsite + 1, the most common residues were R274 and E584 at site 2 and I397 and Q398 at site 4. Based on the data obtained from binding free energy changes (ΔG°_binding), pharmacological property analysis and molecular dynamic simulations, two ChemPLP compounds, 338175 and 1146525, and one RF-Score-VS compound, 337447, were considered as the likely lead compounds. The most promising compound, 1146525, could serve as a scaffold for the future design of novel antimicrobial agents against Vibrio infections.

Interplay of phytohormones and epigenetic regulation: A recipe for plant development and plasticity

Both phytohormone signaling and epigenetic mechanisms have long been known to play crucial roles in plant development and plasticity in response to ambient stimuli. Indeed, diverse signaling pathways mediated by phytohormones and epigenetic processes integrate multiple upstream signals to regulate various plant traits. Emerging evidence indicates that phytohormones and epigenetic processes interact at multiple levels. In this review, we summarize the current knowledge of the interplay between phytohormones and epigenetic processes from the perspective of phytohormone biology. We also review chemical regulators used in epigenetic studies and propose strategies for developing novel regulators using multidisciplinary approaches.

A fully differentiable ligand pose optimization framework guided by deep learning and a traditional scoring function

The recently reported machine learning- or deep learning-based scoring functions (SFs) have shown exciting performance in predicting protein-ligand binding affinities with fruitful application prospects. However, the differentiation between highly similar ligand conformations, including the native binding pose (the global energy minimum state), remains challenging that could greatly enhance the docking. In this work, we propose a fully differentiable, end-to-end framework for ligand pose optimization based on a hybrid SF called DeepRMSD+Vina combined with a multi-layer perceptron (DeepRMSD) and the traditional AutoDock Vina SF. The DeepRMSD+Vina, which combines (1) the root mean square deviation (RMSD) of the docking pose with respect to the native pose and (2) the AutoDock Vina score, is fully differentiable; thus is capable of optimizing the ligand binding pose to the energy-lowest conformation. Evaluated by the CASF-2016 docking power dataset, the DeepRMSD+Vina reaches a success rate of 94.4%, which outperforms most reported SFs to date. We evaluated the ligand conformation optimization framework in practical molecular docking scenarios (redocking and cross-docking tasks), revealing the high potentialities of this framework in drug design and discovery. Structural analysis shows that this framework has the ability to identify key physical interactions in protein-ligand binding, such as hydrogen-bonding. Our work provides a paradigm for optimizing ligand conformations based on deep learning algorithms. The DeepRMSD+Vina model and the optimization framework are available at GitHub repository https://github.com/zchwang/DeepRMSD-Vina_Optimization.

HCovDock: an efficient docking method for modeling covalent protein-ligand interactions

Covalent inhibitors have received extensive attentions in the past few decades because of their long residence time, high binding efficiency and strong selectivity. Therefore, it is valuable to develop computational tools like molecular docking for modeling of covalent protein-ligand interactions or screening of potential covalent drugs. Meeting the needs, we have proposed HCovDock, an efficient docking algorithm for covalent protein-ligand interactions by integrating a ligand sampling method of incremental construction and a scoring function with covalent bond-based energy. Tested on a benchmark containing 207 diverse protein-ligand complexes, HCovDock exhibits a significantly better performance than seven other state-of-the-art covalent docking programs (AutoDock, Cov_DOX, CovDock, FITTED, GOLD, ICM-Pro and MOE). With the criterion of ligand root-mean-squared distance < 2.0 Å, HCovDock obtains a high success rate of 70.5% and 93.2% in reproducing experimentally observed structures for top 1 and top 10 predictions. In addition, HCovDock is also validated in virtual screening against 10 receptors of three proteins. HCovDock is computationally efficient and the average running time for docking a ligand is only 5 min with as fast as 1 sec for ligands with one rotatable bond and about 18 min for ligands with 23 rotational bonds. HCovDock can be freely assessed at http://huanglab.phys.hust.edu.cn/hcovdock/.

Multitargeted Molecular Docking and Dynamic Simulation Studies of Bioactive Compounds from Rosmarinus officinalis against Alzheimer’s Disease

Alzheimer’s disease (AD) has been associated with the hallmark features of cholinergic dysfunction, amyloid beta (Aβ) aggregation and impaired synaptic transmission, which makes the associated proteins, such as β-site amyloid precursor protein cleaving enzyme 1 (BACE I), acetylcholine esterase (AChE) and synapsin I, II and III, major targets for therapeutic intervention. The present study investigated the therapeutic potential of three major phytochemicals of Rosmarinus officinalis, ursolic acid (UA), rosmarinic acid (RA) and carnosic acid (CA), based on their binding affinity with AD-associated proteins. Detailed docking studies were conducted using AutoDock vina followed by molecular dynamic (MD) simulations using Amber 20. The docking analysis of the selected molecules showed the binding energies of their interaction with the target proteins, while MD simulations comprising root mean square deviation (RMSD), root mean square fluctuation (RMSF) and molecular mechanics/generalized born surface area (MM/GBSA) binding free energy calculations were carried out to check the stability of bound complexes. The drug likeness and the pharmacokinetic properties of the selected molecules were also checked through the Lipinski filter and ADMETSAR analysis. All these bioactive compounds demonstrated strong binding affinity with AChE, BACE1 and synapsin I, II and III. The results showed UA and RA to be potential inhibitors of AChE and BACE1, exhibiting binding energies comparable to those of donepezil, used as a positive control. The drug likeness and pharmacokinetic properties of these compounds also demonstrated drug-like characteristics, indicating the need for further in vitro and in vivo investigations to ascertain their therapeutic potential for AD.

New Wine in an Old Bottle: Utilizing Chemical Genetics to Dissect Apical Hook Development

The apical hook is formed by dicot seedlings to protect the tender shoot apical meristem during soil emergence. Regulated by many phytohormones, the apical hook has been taken as a model to study the crosstalk between individual signaling pathways. Over recent decades, the roles of different phytohormones and environmental signals in apical hook development have been illustrated. However, key regulators downstream of canonical hormone signaling have rarely been identified via classical genetics screening, possibly due to genetic redundancy and/or lethal mutation. Chemical genetics that utilize small molecules to perturb and elucidate biological processes could provide a complementary strategy to overcome the limitations in classical genetics. In this review, we summarize current progress in hormonal regulation of the apical hook, and previously reported chemical tools that could assist the understanding of this complex developmental process. We also provide insight into novel strategies for chemical screening and target identification, which could possibly lead to discoveries of new regulatory components in apical hook development, or unidentified signaling crosstalk that is overlooked by classical genetics screening.

Boosting Protein-Ligand Binding Pose Prediction and Virtual Screening Based on Residue-Atom Distance Likelihood Potential and Graph Transformer

The past few years have witnessed enormous progress toward applying machine learning approaches to the development of protein-ligand scoring functions. However, the robust performance and wide applicability of scoring functions remain a big challenge for increasing the success rate of docking-based virtual screening. Herein, a novel scoring function named RTMScore was developed by introducing a tailored residue-based graph representation strategy and several graph transformer layers for the learning of protein and ligand representations, followed by a mixture density network to obtain residue-atom distance likelihood potential. Our approach was resolutely validated on the CASF-2016 benchmark, and the results indicate that RTMScore can outperform almost all of the other state-of-the-art methods in terms of both the docking and screening powers. Further evaluation confirms the robustness of our approach that can not only retain its docking power on cross-docked poses but also achieve improved performance as a rescoring tool in larger-scale virtual screening.

Recent Advances in Application of Computer-Aided Drug Design in Anti-Influenza A Virus Drug Discovery

Influenza A is an acute respiratory infectious disease caused by the influenza A virus, which seriously threatens global human health and causes substantial economic losses every year. With the emergence of new viral strains, anti-influenza drugs remain the most effective treatment for influenza A. Research on traditional, innovative small-molecule drugs faces many challenges, while computer-aided drug design (CADD) offers opportunities for the rapid and effective development of innovative drugs. This literature review describes the general process of CADD, the viral proteins that play an essential role in the life cycle of the influenza A virus and can be used as therapeutic targets for anti-influenza drugs, and examples of drug screening of viral target proteins by applying the CADD approach. Finally, the main limitations of current CADD strategies in anti-influenza drug discovery and the field's future directions are discussed.