Surface ID: a geometry-aware system for protein molecular surface comparison.

MOTIVATION: A protein can be represented in several forms, including its 1D sequence, 3D atom coordinates, and molecular surface. A protein surface contains rich structural and chemical features directly related to the protein's function such as its ability to interact with other molecules. While many methods have been developed for comparing the similarity of proteins using the sequence and structural representations, computational methods based on molecular surface representation are limited. RESULTS: Here, we describe "Surface ID," a geometric deep learning system for high-throughput surface comparison based on geometric and chemical features. Surface ID offers a novel grouping and alignment algorithm useful for clustering proteins by function, visualization, and in silico screening of potential binding partners to a target molecule. Our method demonstrates top performance in surface similarity assessment, indicating great potential for protein functional annotation, a major need in protein engineering and therapeutic design. AVAILABILITY AND IMPLEMENTATION: Source code for the Surface ID model, trained weights, and inference script are available at https://github.com/Sanofi-Public/LMR-SurfaceID.

Autoencoders for dimensionality reduction in molecular dynamics: Collective variable dimension, biasing, and transition states.

The heat shock protein 90 (Hsp90) is a molecular chaperone that controls the folding and activation of client proteins using the free energy of ATP hydrolysis. The Hsp90 active site is in its N-terminal domain (NTD). Our goal is to characterize the dynamics of NTD using an autoencoder-learned collective variable (CV) in conjunction with adaptive biasing force Langevin dynamics. Using dihedral analysis, we cluster all available experimental Hsp90 NTD structures into distinct native states. We then perform unbiased molecular dynamics (MD) simulations to construct a dataset that represents each state and use this dataset to train an autoencoder. Two autoencoder architectures are considered, with one and two hidden layers, respectively, and bottlenecks of dimension k ranging from 1 to 10. We demonstrate that the addition of an extra hidden layer does not significantly improve the performance, while it leads to complicated CVs that increase the computational cost of biased MD calculations. In addition, a two-dimensional (2D) bottleneck can provide enough information of the different states, while the optimal bottleneck dimension is five. For the 2D bottleneck, the 2D CV is directly used in biased MD simulations. For the five-dimensional (5D) bottleneck, we perform an analysis of the latent CV space and identify the pair of CV coordinates that best separates the states of Hsp90. Interestingly, selecting a 2D CV out of the 5D CV space leads to better results than directly learning a 2D CV and allows observation of transitions between native states when running free energy biased dynamics.

Isometric Stratified Ensembles: A Partial and Incremental Adaptive Applicability Domain and Consensus-Based Classification Strategy for Highly Imbalanced Data Sets with Application to Colloidal Aggregation.

Partial and incremental stratification analysis of a quantitative structure-interference relationship (QSIR) is a novel strategy intended to categorize classification provided by machine learning techniques. It is based on a 2D mapping of classification statistics onto two categorical axes: the degree of consensus and level of applicability domain. An internal cross-validation set allows to determine the statistical performance of the ensemble at every 2D map stratum and hence to define isometric local performance regions with the aim of better hit ranking and selection. During training, isometric stratified ensembles (ISE) applies a recursive decorrelated variable selection and considers the cardinal ratio of classes to balance training sets and thus avoid bias due to possible class imbalance. To exemplify the interest of this strategy, three different highly imbalanced PubChem pairs of AmpC ß-lactamase and cruzain inhibition assay campaigns of colloidal aggregators and complementary aggregators data set available at the AGGREGATOR ADVISOR predictor web page were employed. Statistics obtained using this new strategy show outperforming results compared to former published tools, with and without a classical applicability domain. ISE performance on classifying colloidal aggregators shows from a global AUC of 0.82, when the whole test data set is considered, up to a maximum AUC of 0.88, when its highest confidence isometric stratum is retained.

Protein loops with multiple meta-stable conformations: A challenge for sampling and scoring methods.

Flexible regions in proteins, such as loops, cannot be represented by a single conformation. Instead, conformational ensembles are needed to provide a more global picture. In this context, identifying statistically meaningful conformations within an ensemble generated by loop sampling techniques remains an open problem. The difficulty is primarily related to the lack of structural data about these flexible regions. With the majority of structural data coming from x-ray crystallography and ignoring plasticity, the conception and evaluation of loop scoring methods is challenging. In this work, we compare the performance of various scoring methods on a set of eight protein loops that are known to be flexible. The ability of each method to identify and select all of the known conformations is assessed, and the underlying energy landscapes are produced and projected to visualize the qualitative differences obtained when using the methods. Statistical potentials are found to provide considerable reliability despite their being designed to tradeoff accuracy for lower computational cost. On a large pool of loop models, they are capable of filtering out statistically improbable states while retaining those that resemble known (and thus likely) conformations. However, computationally expensive methods are still required for more precise assessment and structural refinement. The results also highlight the importance of employing several scaffolds for the protein, due to the high influence of small structural rearrangements in the rest of the protein over the modeled energy landscape for the loop.

Scaffold-Constrained Molecular Generation.

One of the major applications of generative models for drug discovery targets the lead-optimization phase. During the optimization of a lead series, it is common to have scaffold constraints imposed on the structure of the molecules designed. Without enforcing such constraints, the probability of generating molecules with the required scaffold is extremely low and hinders the practicality of generative models for de novo drug design. To tackle this issue, we introduce a new algorithm, named SAMOA (Scaffold Constrained Molecular Generation), to perform scaffold-constrained in silico molecular design. We build on the well-known SMILES-based Recurrent Neural Network (RNN) generative model, with a modified sampling procedure to achieve scaffold-constrained generation. We directly benefit from the associated reinforcement learning methods, allowing to design molecules optimized for different properties while exploring only the relevant chemical space. We showcase the method's ability to perform scaffold-constrained generation on various tasks: designing novel molecules around scaffolds extracted from SureChEMBL chemical series, generating novel active molecules on the Dopamine Receptor D2 (DRD2) target, and finally, designing predicted actives on the MMP-12 series, an industrial lead-optimization project.

Cytoplasmic STAT3 represses autophagy by inhibiting PKR activity.

In a screen designed to identify novel inducers of autophagy, we discovered that STAT3 inhibitors potently stimulate the autophagic flux. Accordingly, genetic inhibition of STAT3 stimulated autophagy in vitro and in vivo, while overexpression of STAT3 variants, encompassing wild-type, nonphosphorylatable, and extranuclear STAT3, inhibited starvation-induced autophagy. The SH2 domain of STAT3 was found to interact with the catalytic domain of the eIF2α kinase 2 EIF2AK2, best known as protein kinase R (PKR). Pharmacological and genetic inhibition of STAT3 stimulated the activating phosphorylation of PKR and consequent eIF2α hyperphosphorylation. Moreover, PKR depletion inhibited autophagy as initiated by chemical STAT3 inhibitors or free fatty acids like palmitate. STAT3-targeting chemicals and palmitate caused the disruption of inhibitory STAT3-PKR interactions, followed by PKR-dependent eIF2α phosphorylation, which facilitates autophagy induction. These results unravel an unsuspected mechanism of autophagy control that involves STAT3 and PKR as interacting partners.

Differential Water Thermodynamics Determine PI3K-Beta/Delta Selectivity for Solvent-Exposed Ligand Modifications.

Phosphoinositide 3-kinases (PI3Ks) are involved in important cellular functions and represent desirable targets for drug discovery efforts, especially related to oncology; however, the four PI3K subtypes (α, ß, γ, and δ) have highly similar binding sites, making the design of selective inhibitors challenging. A series of inhibitors with selectivity toward the ß subtype over δ resulted in compound 3(S), which has entered a phase I/Ib clinical trial for patients with advanced PTEN-deficient cancer. Interestingly, X-ray crystallography revealed that the modifications making inhibitor 3(S) and related compounds selective toward the ß-isoform do not interact directly with either PI3Kß or PI3Kδ, thereby confounding rationalization of the SAR. Here, we apply explicit solvent molecular dynamics and solvent thermodynamic analysis using WaterMap in an effort to understand the unusual affinity and selectivity trends. We find that differences in solvent energetics and water networks, which are modulated upon binding of different ligands, explain the experimental affinity and selectivity trends. This study highlights the critical role of water molecules in molecular recognition and the importance of considering water networks in drug discovery efforts to rationalize and improve selectivity.

Engineering protein therapeutics: predictive performances of a structure-based virtual affinity maturation protocol.

The implementation of a structure based virtual affinity maturation protocol and evaluation of its predictivity are presented. The in silico protocol is based on conformational sampling of the interface residues (using the Dead End Elimination/A* algorithm), followed by the estimation of the change of free energy of binding due to a point mutation, applying MM/PBSA calculations. Several implementations of the protocol have been evaluated for 173 mutations in 7 different protein complexes for which experimental data were available: the use of the Boltzamnn averaged predictor based on the free energy of binding (ΔΔG(*)) combined with the one based on its polar component only (ΔΔE(pol*)) led to the proposal of a subset of mutations out of which 45% would have successfully enhanced the binding. When focusing on those mutations that are less likely to be introduced by natural in vivo maturation methods (99 mutations with at least two base changes in the codon), the success rate is increased to 63%. In another evaluation, focusing on 56 alanine scanning mutations, the in silico protocol was able to detect 89% of the hot-spots.

Contact Map Fingerprints of Protein-Ligand Unbinding Trajectories Reveal Mechanisms Determining Residence Times Computed from Scaled Molecular Dynamics.

The binding kinetic properties of potential drugs may significantly influence their subsequent clinical efficacy. Predictions of these properties based on computer simulations provide a useful alternative to their expensive and time-consuming experimental counterparts, even at an early drug discovery stage. Herein, we perform scaled molecular dynamics (ScaledMD) simulations on a set of 27 ligands of HSP90 belonging to more than seven chemical series to estimate their relative residence times. We introduce two new techniques for the analysis and the classification of the simulated unbinding trajectories. The first technique, which helps in estimating the limits of the free energy well around the bound state, and the second one, based on a new contact map fingerprint, allow the description and the comparison of the paths that lead to unbinding. Using these analyses, we find that ScaledMD's relative residence time generally enables the identification of the slowest unbinders. We propose an explanation for the underestimation of the residence times of a subset of compounds, and we investigate how the biasing in ScaledMD can affect the mechanistic insights that can be gained from the simulations.

Machine Learning Force Fields and Coarse-Grained Variables in Molecular Dynamics: Application to Materials and Biological Systems.

Machine learning encompasses tools and algorithms that are now becoming popular in almost all scientific and technological fields. This is true for molecular dynamics as well, where machine learning offers promises of extracting valuable information from the enormous amounts of data generated by simulation of complex systems. We provide here a review of our current understanding of goals, benefits, and limitations of machine learning techniques for computational studies on atomistic systems, focusing on the construction of empirical force fields from ab initio databases and the determination of reaction coordinates for free energy computation and enhanced sampling.

Conformational changes in antibody Fab fragments upon binding and their consequences on the performance of docking algorithms.

BACKGROUND: The existence of conformational changes in antibodies upon binding has been previously established. However, existing analyses focus on individual cases and no quantitative study provides a more global view of potential moves and repacking, especially on recent data. The present study focuses on analyzing the conformational changes in various antibodies upon binding, providing quantitative observations to be exploited for antibody-related modeling. METHODS: Cartesian and dihedral Root-Mean-Squared Deviations were calculated for different subparts of 27 different antibodies, for which X-ray structures in the bound and unbound states are available. Elbow angle variations were also calculated. Previously reported results of four docking algorithms were condensed into one score giving overall docking success for each of 16 antibody-antigen cases. RESULTS: Very diverse movements are observed upon binding. While many loops stay very rigid, several others display side-chain repacking or backbone rearrangements, or both, at many different levels. Large conformational changes restricted to one or more antibody hypervariable loops were found to be a better indicator of docking difficulty than overall conformational variation at the antibody-antigen interface. However, the failure of docking algorithms on some almost-rigid cases shows that scoring is still a major bottleneck in docking pose prediction. CONCLUSIONS: This study is aimed to help scientists working on antibody analysis and design by giving insights into the nature and the extent of conformational changes at different levels upon antigen binding.

SAR156497, an exquisitely selective inhibitor of aurora kinases.

The Aurora family of serine/threonine kinases is essential for mitosis. Their crucial role in cell cycle regulation and aberrant expression in a broad range of malignancies have been demonstrated and have prompted intensive search for small molecule Aurora inhibitors. Indeed, over 10 of them have reached the clinic as potential anticancer therapies. We report herein the discovery and optimization of a novel series of tricyclic molecules that has led to SAR156497, an exquisitely selective Aurora A, B, and C inhibitor with in vitro and in vivo efficacy. We also provide insights into its mode of binding to its target proteins, which could explain its selectivity.

An electrostatic basis for the stability of thermophilic proteins.

Two factors provide key contributions to the stability of thermophilic proteins relative to their mesophilic homologues: electrostatic interactions of charged residues in the folded state and the dielectric response of the folded protein. The dielectric response for proteins in a "thermophilic series" globally modulates the thermal stability of its members, with the calculated dielectric constant for the protein increasing from mesophiles to hyperthermophiles. This variability results from differences in the distribution of charged residues on the surface of the protein, in agreement with structural and genetic observations. Furthermore, the contribution of electrostatic interactions to the stability of the folded state is more favorable for thermophilic proteins than for their mesophilic homologues. This leads to the conclusion that electrostatic interactions play an important role in determining the stability of proteins at high temperatures. The interplay between electrostatic interactions and dielectric response also provides further rationalization for the enhanced stability of thermophilic proteins with respect to cold-denaturation. Taken together, the distribution of charged residues and their fluctuations have been shown to be factors in modulating protein stability over the entire range of biologically relevant temperatures.

Tricyclic series of heat shock protein 90 (Hsp90) inhibitors part I: discovery of tricyclic imidazo[4,5-c]pyridines as potent inhibitors of the Hsp90 molecular chaperone.

A novel class of heat shock protein 90 (Hsp90) inhibitors was developed after a low throughput screen (LTS) of a focused library containing approximately 21K compounds selected by virtual screening. The initial [1-{3-H-imidazo[4-5-c]pyridin-2-yl}-3,4-dihydro-2H-pyrido[2,1-a]isoindole-6-one] (1) compound showed moderate activity (IC(50) = 7.6 µM on Hsp82, the yeast homologue of Hsp90). A high-resolution X-ray structure shows that compound 1 binds into an "induced" hydrophobic pocket, 10-15 Å away from the ATP/resorcinol binding site. Iterative cycles of structure-based drug design (SBDD) and chemical synthesis led to the design and preparation of analogues with improved affinity. These optimized molecules make productive interactions within the ATP binding site as reported by other Hsp90 inhibitors. This resulted in compound 8, which is a highly potent inhibitor in biochemical and cellular assays (K(d) = 0.35 nM on Hsp90; IC(50) = 30 nM on SKBr3 mammary carcinoma cells) and in an in vivo leukemia model.