Krzysztof Rataj

Impact of Template Choice on Homology Model Efficiency in Virtual Screening

Homology modeling is a reliable method of predicting the three-dimensional structures of proteins that lack NMR or X-ray crystallographic data. It employs the assumption that a structural resemblance exists between closely related proteins. Despite the availability of many crystal structures of possible templates, only the closest ones are chosen for homology modeling purposes. To validate the aforementioned approach, we performed homology modeling of four serotonin receptors (5-HT1AR, 5-HT2AR, 5-HT6R, 5-HT7R) for virtual screening purposes, using 10 available G-Protein Coupled Receptors (GPCR) templates with diverse evolutionary distances to the targets, with various approaches to alignment construction and model building. The resulting models were further validated in two steps by means of ligand docking and enrichment calculation, using Glide software. The final quality of the models was determined in virtual screening-like experiments by the AUROC score of the resulting ROC curves. The outcome of this research showed that no correlation between sequence identity and model quality was found, leading to the conclusion that the closest phylogenetic relative is not always the best template for homology modeling.

An application of machine learning methods to structural interaction fingerprints—a case study of kinase inhibitors

In this Letter, we present a novel methodology of searching for biologically active compounds, which is based on the combination of docking experiments and analysis of the results by machine learning methods. The study was performed for 5 different protein kinases, and several sets of compounds (active, inactive and assumed inactives) were docked into their targets. The resulting ligand-protein complexes were represented by the means of structural interaction fingerprints profiles (SIFts profiles) that constituted an input for ML methods. The developed protocol was found to be superior to the combination of classification algorithms with the standard fingerprint MACCSFP.

Multi-Step Protocol for Automatic Evaluation of Docking Results Based on Machine Learning Methods—A Case Study of Serotonin Receptors 5-HT6 and 5-HT7

Molecular docking, despite its undeniable usefulness in computer-aided drug design protocols and the increasing sophistication of tools used in the prediction of ligand-protein interaction energies, is still connected with a problem of effective results analysis. In this study, a novel protocol for the automatic evaluation of numerous docking results is presented, being a combination of Structural Interaction Fingerprints and Spectrophores descriptors, machine-learning techniques, and multi-step results analysis. Such an approach takes into consideration the performance of a particular learning algorithm (five machine learning methods were applied), the performance of the docking algorithm itself, the variety of conformations returned from the docking experiment, and the receptor structure (homology models were constructed on five different templates). Evaluation using compounds active toward 5-HT6 and 5-HT7 receptors, as well as additional analysis carried out for beta-2 adrenergic receptor ligands, proved that the methodology is a viable tool for supporting virtual screening protocols, enabling proper discrimination between active and inactive compounds.

From Homology Models to a Set of Predictive Binding Pockets–a 5-HT1A Receptor Case Study

Despite its remarkable importance in the arena of drug design, serotonin 1A receptor (5-HT1A) has been elusive to the X-ray crystallography community. This lack of direct structural information not only hampers our knowledge regarding the binding modes of many popular ligands (including the endogenous neurotransmitter-serotonin), but also limits the search for more potent compounds. In this paper we shed new light on the 3D pharmacological properties of the 5-HT1A receptor by using a ligand-guided approach (ALiBERO) grounded in the Internal Coordinate Mechanics (ICM) docking platform. Starting from a homology template and set of known actives, the method introduces receptor flexibility via Normal Mode Analysis and Monte Carlo sampling, to generate a subset of pockets that display enriched discrimination of actives from inactives in retrospective docking. Here, we thoroughly investigated the repercussions of using different protein templates and the effect of compound selection on screening performance. Finally, the best resulting protein models were applied prospectively in a large virtual screening campaign, in which two new active compounds were identified that were chemically distinct from those described in the literature.

Multiple conformational states in retrospective virtual screening – homology models vs. crystal structures: beta-2 adrenergic receptor case study

AbstractBackgroundDistinguishing active from inactive compounds is one of the crucial problems of molecular docking, especially in the context of virtual screening experiments. The randomization of poses and the natural flexibility of the protein make this discrimination even harder. Some of the recent approaches to post-docking analysis use an ensemble of receptor models to mimic this naturally occurring conformational diversity. However, the optimal number of receptor conformations is yet to be determined.In this study, we compare the results of a retrospective screening of beta-2 adrenergic receptor ligands performed on both the ensemble of receptor conformations extracted from ten available crystal structures and an equal number of homology models. Additional analysis was also performed for homology models with up to 20 receptor conformations considered.ResultsThe docking results were encoded into the Structural Interaction Fingerprints and were automatically analyzed by support vector machine. The use of homology models in such virtual screening application was proved to be superior in comparison to crystal structures. Additionally, increasing the number of receptor conformational states led to enhanced effectiveness of active vs. inactive compounds discrimination.ConclusionsFor virtual screening purposes, the use of homology models was found to be most beneficial, even in the presence of crystallographic data regarding the conformational space of the receptor. The results also showed that increasing the number of receptors considered improves the effectiveness of identifying active compounds by machine learning methods. Graphical abstractComparison of machine learning results obtained for various number of beta-2 AR homology models and crystal structures.

The importance of template choice in homology modeling. A 5-HT6R case study

Over the years, homology modelling has grown into an important tool for biochemistry and pharmacology. It allows prediction of three-dimensional structure of proteins, which have not been resolved with empirical methods. The final outcome of such research is affected by many factors, the choice of template being a crucial one. Current paradigm states, that proteins with the smallest evolutionary distance and thus, the highest identity/similarity, to the target, should achieve the highest performance. The goal of this research was to verify the credibility of this assumption when incorporating homology models into Virtual Screening protocol. The target for this case study is 5-HT6R, which belongs to class A GPCR, and as a trans-membrane protein, is extremely hard to crystallize or solubilize. This makes standard protein structure assessment inexplicably difficult, however a few members of class A GPCR had their structure solved. 5-HT6R itself is involved in learning, memorizing and overall cognition processes, and is a target in anti-depression drug research [1,2]. This study comprised of homology modelling of 5-HT6R based on seven available GPCR templates (A2A, beta1, beta2, CXCR4, D3, H1, rhodopsin), and further verification of created models by means of ligand docking (Glide). The quality of generated structures was assessed in three subsequent steps, each consisting of different compounds sets for docking procedure. The final models were selected basing on the number of active ligands to decoys docked ratio. Interestingly, the templates used to construct the most successful models were not the evolutionarily closest ones, therefore putting the existing paradigm into question when it comes to VS application.

Application of Structural Interaction Fingerpints (SIFts) into post-docking analysis - insight into activity and selectivity

Cheminformatic methods, such as Virtual Screening, constitute a vital part of modern drug design process. This technique enables not only viable prediction of physicochemical properties of the molecules, but also effective database mining, being particularly useful tool in search for ligands of desired activity. Successful performance in case of single target drugs, implies a potential to extend its capabilities to compounds bearing desired activity towards multiple receptors. In this research, we present application of Structural Interaction Fingerprints (SIFts) [1] combined with Machine Learning (ML) as a method to select single- and multi-target ligands from the docking results. A handful of protein kinases pairs was designated as targets. Collection of pseudo selective compounds, with various activity profiles, was acquired from ChEMBL database. Furthermore, for each target a set of ligands of various activity was aggregated. Decoy structures were random ligands from ZINC database. The compounds were docked into respective proteins, and SIFts were calculated for each protein-ligand complex. Training sets used in ML experiments consisted of cluster centroids of active and inactive ligands, whereas test sets were composed of remaining compounds, that returned docking poses. The key aim of this study is to develop a viable method to filter the docking results, so that the compounds meeting desired activity profile are selected.

Fingerprint-Based Machine Learning Approach to Identify Potent and Selective 5-HT2BR Ligands

The identification of subtype-selective GPCR (G-protein coupled receptor) ligands is a challenging task. In this study, we developed a computational protocol to find compounds with 5-HT2BR versus 5-HT1BR selectivity. Our approach employs the hierarchical combination of machine learning methods, docking, and multiple scoring methods. First, we applied machine learning tools to filter a large database of druglike compounds by the new Neighbouring Substructures Fingerprint (NSFP). This two-dimensional fingerprint contains information on the connectivity of the substructural features of a compound. Preselected subsets of the database were then subjected to docking calculations. The main indicators of compounds’ selectivity were their different interactions with the secondary binding pockets of both target proteins, while binding modes within the orthosteric binding pocket were preserved. The combined methodology of ligand-based and structure-based methods was validated prospectively, resulting in the identification of hits with nanomolar affinity and ten-fold to ten thousand-fold selectivities.

Substructural Connectivity Fingerprint and Extreme Entropy Machines—A New Method of Compound Representation and Analysis

Key-based substructural fingerprints are an important element of computer-aided drug design techniques. The usefulness of the fingerprints in filtering compound databases is invaluable, as they allow for the quick rejection of molecules with a low probability of being active. However, this method is flawed, as it does not consider the connections between substructures. After changing the connections between particular chemical moieties, the fingerprint representation of the compound remains the same, which leads to difficulties in distinguishing between active and inactive compounds. In this study, we present a new method of compound representation—substructural connectivity fingerprints (SCFP), providing information not only about the presence of particular substructures in the molecule but also additional data on substructure connections. Such representation was analyzed by the recently developed methodology—extreme entropy machines (EEM). The SCFP can be a valuable addition to virtual screening tools, as it represents compound structure with greater detail and more specificity, allowing for more accurate classification.

The salt bridge in ligand-protein complexes - systematic theoretical and statistical investigations

Although the salt bridge is the strongest among all known noncovalent molecular interactions, no comprehensive studies have been conducted to date to examine its role and significance in drug design. Thus, a systematic study of the salt bridge in biological systems is reported herein, with a broad analysis of publicly available data from Protein Data Bank, DrugBank, ChEMBL, and GPCRdb. The results revealed the distance and angular preferences as well as privileged molecular motifs of salt bridges in ligand-receptor complexes, which could be used to design the strongest interactions. Moreover, using quantum chemical calculations at the MP2 level, the energetic, directionality, and spatial variabilities of salt bridges were investigated using simple model systems mimicking salt bridges in a biological environment. Additionally, natural orbitals for chemical valence (NOCV) combined with the extended-transition-state (ETS) bond-energy decomposition method (ETS-NOCV) were analyzed and indicated a strong covalent contribution to the salt bridge interaction. The present results could be useful for implementation in rational drug design protocols.