Convex-PL: a novel knowledge-based potential for protein-ligand interactions deduced from structural databases using convex optimization

Identification of a Clade-Specific HLA-C*03:02 CTL Epitope GY9 Derived from the HIV-1 p17 Matrix Protein

Efforts towards an effective HIV-1 vaccine have remained mainly unsuccessful. There is increasing evidence for a potential role of HLA-C-restricted CD8+ T cell responses in HIV-1 control, including our recent report of HLA-C*03:02 among African children. However, there are no documented optimal HIV-1 CD8+ T cell epitopes restricted by HLA-C*03:02; additionally, the structural influence of HLA-C*03:02 on epitope binding is undetermined. Immunoinformatics approaches provide a fast and inexpensive method to discover HLA-restricted epitopes. Here, we employed immunopeptidomics to identify HLA-C*03:02 CD8+ T cell epitopes. We identified a clade-specific Gag-derived GY9 (GTEELRSLY) HIV-1 p17 matrix epitope potentially restricted to HLA-C*03:02. Residues E62, T142, and E151 in the HLA-C*03:02 binding groove and positions p3, p6, and p9 on the GY9 epitope are crucial in shaping and stabilizing the epitope binding. Our findings support the growing evidence of the contribution of HLA-C molecules to HIV-1 control and provide a prospect for vaccine strategies.

Drugging the entire human proteome: Are we there yet?

Each of the ∼20,000 proteins in the human proteome is a potential target for compounds that bind to it and modify its function. The 3D structures of most of these proteins are now available. Here, we discuss the prospects for using these structures to perform proteome-wide virtual HTS (VHTS). We compare physics-based (docking) and AI VHTS approaches, some of which are now being applied with large databases of compounds to thousands of targets. Although preliminary proteome-wide screens are now within our grasp, further methodological developments are expected to improve the accuracy of the results.

Arylureidoaurones: Synthesis, in vitro α-glucosidase, and α-amylase inhibition activity

Because of the colossal global burden of diabetes, there is an urgent need for more effective and safer drugs. We designed and synthesized a new series of aurone derivatives possessing phenylureido or bis-phenylureido moieties as α-glucosidase and α-amylase inhibitors. Most of the synthesized phenylureidoaurones have demonstrated superior inhibition activities (IC50s of 9.6-339.9 μM) against α-glucosidase relative to acarbose (IC50 = 750.0 μM) as the reference drug. Substitution of aurone analogues with two phenylureido substituents at the 5-position of the benzofuranone moiety and the 3' or 4' positions of the 2-phenyl ring resulted in compounds with almost 120-180 times more potent inhibitory activities than acarbose. The aurone analogue possessing two phenylureido substitutions at 5 and 4' positions (13) showed the highest inhibition activity with an IC50 of 4.2 ± 0.1 μM. Kinetic studies suggested their inhibition mode to be competitive. We also investigated the binding mode of the most potent compounds using the consensually docked 4D-QSAR methodology. Furthermore, these analogues showed weak-to-moderate non-competitive inhibitory activity against α-amylase. 5-Methyl substituted aurone with 4'-phenylureido moiety (6e) demonstrated the highest inhibition activity on α-amylase with an IC50 of 142.0 ± 1.6 μM relative to acarbose (IC50 = 108 ± 1.2 μM). Our computational studies suggested that these analogues interact with a hydrophilic allosteric site in α-amylase, located far from the enzyme active site at the N-terminal.

Elucidating the molecular mechanisms of essential oils' insecticidal action using a novel cheminformatics protocol

Essential oils (EOs) are a promising source for novel environmentally safe insecticides. However, the structural diversity of their compounds poses challenges to accurately elucidate their biological mechanisms of action. We present a new chemoinformatics methodology aimed at predicting the impact of essential oil (EO) compounds on the molecular targets of commercial insecticides. Our approach merges virtual screening, chemoinformatics, and machine learning to identify custom signatures and reference molecule clusters. By assigning a molecule to a cluster, we can determine its most likely interaction targets. Our findings reveal that the main targets of EOs are juvenile hormone-specific proteins (JHBP and MET) and octopamine receptor agonists (OctpRago). Three of the twenty clusters show strong similarities to the juvenile hormone, steroids, and biogenic amines. For instance, the methodology successfully identified E-Nerolidol, for which literature points indications of disrupting insect metamorphosis and neurochemistry, as a potential insecticide in these pathways. We validated the predictions through experimental bioassays, observing symptoms in blowflies that were consistent with the computational results. This new approach sheds a higher light on the ways of action of EO compounds in nature and biotechnology. It also opens new possibilities for understanding how molecules can interfere with biological systems and has broad implications for areas such as drug design.

Assessment of the Generalization Abilities of Machine-Learning Scoring Functions for Structure-Based Virtual Screening

In structure-based virtual screening (SBVS), it is critical that scoring functions capture protein-ligand atomic interactions. By focusing on the local domains of ligand binding pockets, a standardized pocket Pfam-based clustering (Pfam-cluster) approach was developed to assess the cross-target generalization ability of machine-learning scoring functions (MLSFs). Subsequently, 12 typical MLSFs were evaluated using random cross-validation (Random-CV), protein sequence similarity-based cross-validation (Seq-CV), and pocket Pfam-based cross-validation (Pfam-CV) methods. Surprisingly, all of the tested models showed decreased performances from Random-CV to Seq-CV to Pfam-CV experiments, not showing satisfactory generalization capacity. Our interpretable analysis suggested that the predictions on novel targets by MLSFs were dependent on buried solvent-accessible surface area (SASA)-related features of complex structures, with greater predicted binding affinities on complexes owning larger protein-ligand interfaces. By combining buried SASA-related features with target-specific patterns that were only shared among structurally similar compounds in the same cluster, the random forest (RF)-Score attained a good performance in the Random-CV test. Based on these findings, we strongly advise assessing the generalization ability of MLSFs with the Pfam-cluster approach and being cautious with the features learned by MLSFs.

Scoring Functions for Protein-Ligand Binding Affinity Prediction using Structure-Based Deep Learning: A Review

The rapid and accurate in silico prediction of protein-ligand binding free energies or binding affinities has the potential to transform drug discovery. In recent years, there has been a rapid growth of interest in deep learning methods for the prediction of protein-ligand binding affinities based on the structural information of protein-ligand complexes. These structure-based scoring functions often obtain better results than classical scoring functions when applied within their applicability domain. Here we review structure-based scoring functions for binding affinity prediction based on deep learning, focussing on different types of architectures, featurization strategies, data sets, methods for training and evaluation, and the role of explainable artificial intelligence in building useful models for real drug-discovery applications.

Using diverse potentials and scoring functions for the development of improved machine-learned models for protein-ligand affinity and docking pose prediction

The advent of computational drug discovery holds the promise of significantly reducing the effort of experimentalists, along with monetary cost. More generally, predicting the binding of small organic molecules to biological macromolecules has far-reaching implications for a range of problems, including metabolomics. However, problems such as predicting the bound structure of a protein-ligand complex along with its affinity have proven to be an enormous challenge. In recent years, machine learning-based methods have proven to be more accurate than older methods, many based on simple linear regression. Nonetheless, there remains room for improvement, as these methods are often trained on a small set of features, with a single functional form for any given physical effect, and often with little mention of the rationale behind choosing one functional form over another. Moreover, it is not entirely clear why one machine learning method is favored over another. In this work, we endeavor to undertake a comprehensive effort towards developing high-accuracy, machine-learned scoring functions, systematically investigating the effects of machine learning method and choice of features, and, when possible, providing insights into the relevant physics using methods that assess feature importance. Here, we show synergism among disparate features, yielding adjusted R² with experimental binding affinities of up to 0.871 on an independent test set and enrichment for native bound structures of up to 0.913. When purely physical terms that model enthalpic and entropic effects are used in the training, we use feature importance assessments to probe the relevant physics and hopefully guide future investigators working on this and other computational chemistry problems.

KORP-PL: a coarse-grained knowledge-based scoring function for protein-ligand interactions

MOTIVATION

Despite the progress made in studying protein-ligand interactions and the widespread application of docking and affinity prediction tools, improving their precision and efficiency still remains a challenge. Computational approaches based on the scoring of docking conformations with statistical potentials constitute a popular alternative to more accurate but costly physics-based thermodynamic sampling methods. In this context, a minimalist and fast sidechain-free knowledge-based potential with a high docking and screening power can be very useful when screening a big number of putative docking conformations.

RESULTS

Here, we present a novel coarse-grained potential defined by a 3D joint probability distribution function that only depends on the pairwise orientation and position between protein backbone and ligand atoms. Despite its extreme simplicity, our approach yields very competitive results with the state-of-the-art scoring functions, especially in docking and screening tasks. For example, we observed a twofold improvement in the median 5% enrichment factor on the DUD-E benchmark compared to Autodock Vina results. Moreover, our results prove that a coarse sidechain-free potential is sufficient for a very successful docking pose prediction.

AVAILABILITYAND IMPLEMENTATION

The standalone version of KORP-PL with the corresponding tests and benchmarks are available at https://team.inria.fr/nano-d/korp-pl/ and https://chaconlab.org/modeling/korp-pl.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Identification of First-in-Class Inhibitors of Kallikrein-Related Peptidase 6 That Promote Oligodendrocyte Differentiation

Multiple sclerosis (MS) is an autoimmune demyelinating disease of the central nervous system (CNS) that causes severe motor, sensory, and cognitive impairments. Kallikrein-related peptidase (KLK)6 is the most abundant serine protease secreted in the CNS, mainly by oligodendrocytes, the myelin-producing cells of the CNS, and KLK6 is assumed to be a robust biomarker of MS, since it is highly increased in the cerebrospinal fluid (CSF) of MS patients. Here, we report the design and biological evaluation of KLK6's low-molecular-weight inhibitors, para-aminobenzyl derivatives. Interestingly, selected hit compounds were selective of the KLK6 proteolytic network encompassing KLK1 and plasmin that also participate in the development of MS physiopathology. Moreover, hits were found noncytotoxic on primary cultures of murine neurons and oligodendrocyte precursor cells (OPCs). Among them, two compounds (32 and 42) were shown to promote the differentiation of OPCs into mature oligodendrocytes in vitro constituting thus emerging leads for the development of regenerative therapies.

Isoform-Selective Enzyme Inhibitors by Exploring Pocket Size According to the Lock-and-Key Principle

In the design of high-affinity and enzyme isoform-selective inhibitors, we applied an approach of augmenting the substituents attached to the benzenesulfonamide scaffold in three ways, namely, substitutions at the 3,5- or 2,4,6-positions or expansion of the condensed ring system. The increased size of the substituents determined the spatial limitations of the active sites of the 12 catalytically active human carbonic anhydrase (CA) isoforms until no binding was observed because of the inability of the compounds to fit in the active site. This approach led to the discovery of high-affinity and high-selectivity compounds for the anticancer target CA IX and antiobesity target CA VB. The x-ray crystallographic structures of compounds bound to CA IX showed the positions of the bound compounds, whereas computational modeling confirmed that steric clashes prevent the binding of these compounds to other isoforms and thus avoid undesired side effects. Such an approach, based on the Lock-and-Key principle, could be used for the development of enzyme-specific drug candidate compounds.

Hydroxylation of Antitubercular Drug Candidate, SQ109, by Mycobacterial Cytochrome P450

Spreading of the multidrug-resistant (MDR) strains of the one of the most harmful pathogen Mycobacterium tuberculosis (Mtb) generates the need for new effective drugs. SQ109 showed activity against resistant Mtb and already advanced to Phase II/III clinical trials. Fast SQ109 degradation is attributed to the human liver Cytochrome P450s (CYPs). However, no information is available about interactions of the drug with Mtb CYPs. Here, we show that Mtb CYP124, previously assigned as a methyl-branched lipid monooxygenase, binds and hydroxylates SQ109 in vitro. A 1.25 Å-resolution crystal structure of the CYP124-SQ109 complex unambiguously shows two conformations of the drug, both positioned for hydroxylation of the ω-methyl group in the trans position. The hydroxylated SQ109 presumably forms stabilizing H-bonds with its target, Mycobacterial membrane protein Large 3 (MmpL3). We anticipate that Mtb CYPs could function as analogs of drug-metabolizing human CYPs affecting pharmacokinetics and pharmacodynamics of antitubercular (anti-TB) drugs.

graphDelta: MPNN Scoring Function for the Affinity Prediction of Protein-Ligand Complexes

In this work, we present graph-convolutional neural networks for the prediction of binding constants of protein-ligand complexes. We derived the model using multi task learning, where the target variables are the dissociation constant (K d), inhibition constant (K i), and half maximal inhibitory concentration (IC50). Being rigorously trained on the PDBbind dataset, the model achieves the Pearson correlation coefficient of 0.87 and the RMSE value of 1.05 in pK units, outperforming recently developed 3D convolutional neural network model K deep.

visGReMLIN: graph mining-based detection and visualization of conserved motifs at 3D protein-ligand interface at the atomic level

Background

Interactions between proteins and non-proteic small molecule ligands play important roles in the biological processes of living systems. Thus, the development of computational methods to support our understanding of the ligand-receptor recognition process is of fundamental importance since these methods are a major step towards ligand prediction, target identification, lead discovery, and more. This article presents visGReMLIN, a web server that couples a graph mining-based strategy to detect motifs at the protein-ligand interface with an interactive platform to visually explore and interpret these motifs in the context of protein-ligand interfaces.

Results

To illustrate the potential of visGReMLIN, we conducted two cases in which our strategy was compared with previous experimentally and computationally determined results. visGReMLIN allowed us to detect patterns previously documented in the literature in a totally visual manner. In addition, we found some motifs that we believe are relevant to protein-ligand interactions in the analyzed datasets.

Conclusions

We aimed to build a visual analytics-oriented web server to detect and visualize common motifs at the protein-ligand interface. visGReMLIN motifs can support users in gaining insights on the key atoms/residues responsible for protein-ligand interactions in a dataset of complexes.

Sampling and refinement protocols for template-based macrocycle docking: 2018 D3R Grand Challenge 4

We describe a new template-based method for docking flexible ligands such as macrocycles to proteins. It combines Monte-Carlo energy minimization on the manifold, a fast manifold search method, with BRIKARD for complex flexible ligand searching, and with the MELD accelerator of Replica-Exchange Molecular Dynamics simulations for atomistic degrees of freedom. Here we test the method in the Drug Design Data Resource blind Grand Challenge competition. This method was among the best performers in the competition, giving sub-angstrom prediction quality for the majority of the targets.

Docking rigid macrocycles using Convex-PL, AutoDock Vina, and RDKit in the D3R Grand Challenge 4

The D3R Grand Challenge 4 provided a brilliant opportunity to test macrocyclic docking protocols on a diverse high-quality experimental data. We participated in both pose and affinity prediction exercises. Overall, we aimed to use an automated structure-based docking pipeline built around a set of tools developed in our team. This exercise again demonstrated a crucial importance of the correct local ligand geometry for the overall success of docking. Starting from the second part of the pose prediction stage, we developed a stable pipeline for sampling macrocycle conformers. This resulted in the subangstrom average precision of our pose predictions. In the affinity prediction exercise we obtained average results. However, we could improve these when using docking poses submitted by the best predictors. Our docking tools including the Convex-PL scoring function are available at https://team.inria.fr/nano-d/software/.

Nonparametric chemical descriptors for the calculation of ligand-biopolymer affinities with machine-learning scoring functions

The computational prediction of ligand-biopolymer affinities is a crucial endeavor in modern drug discovery and one that still poses major challenges. The choice of the appropriate computational method often reveals itself as a trade-off between accuracy and speed, with mathematical devices referred to as scoring functions being the fastest. Among the many shortcomings of scoring functions there is the lack of universal applicability to every molecular system. This is so largely due to their reliance on atom type perception and/or parametrization. This article proposes the use of nonparametric Model of Effective Radii of Atoms descriptors that can be readily computed for the entire Periodic Table and demonstrate that, in combination with machine learning algorithms, they can yield competitive performances and chemically meaningful insights.

An Overview of Scoring Functions Used for Protein-Ligand Interactions in Molecular Docking

Currently, molecular docking is becoming a key tool in drug discovery and molecular modeling applications. The reliability of molecular docking depends on the accuracy of the adopted scoring function, which can guide and determine the ligand poses when thousands of possible poses of ligand are generated. The scoring function can be used to determine the binding mode and site of a ligand, predict binding affinity and identify the potential drug leads for a given protein target. Despite intensive research over the years, accurate and rapid prediction of protein-ligand interactions is still a challenge in molecular docking. For this reason, this study reviews four basic types of scoring functions, physics-based, empirical, knowledge-based, and machine learning-based scoring functions, based on an up-to-date classification scheme. We not only discuss the foundations of the four types scoring functions, suitable application areas and shortcomings, but also discuss challenges and potential future study directions.