Rational Design Of Compounds Research Articles

Molecular insights into degron recognition by CRL5ASB7 ubiquitin ligase

The ankyrin (ANK) SOCS box (ASB) family, encompassing ASB1–18, is the largest group of substrate receptors of cullin 5 Ring E3 ubiquitin ligase. Nonetheless, the mechanism of substrate recognition by ASB family proteins has remained largely elusive. Here we present the crystal structure of ASB7-Elongin B-Elongin C ternary complex bound to a conserved helical degron. ASB7 employs its ANK3-6 to form an extended groove, effectively interacting with the internal α-helix-degron through a network of side-chain-mediated electrostatic and hydrophobic interactions. Our structural findings, combined with biochemical and cellular analyses, identify the key residues of the degron motif and ASB7 required for their recognition. This will facilitate the identification of additional physiological substrates of ASB7 by providing a defined degron motif for screening. Furthermore, the structural insights provide a basis for the rational design of compounds that can specifically target ASB7 by disrupting its interaction with its cognate degron.

Development of Degraders of Cyclin-Dependent Kinases 4 and 6 Based on Rational Drug Design.

Degradation of target proteins has been considered to be a promising therapeutic approach, but the rational design of compounds for degradation remains a challenge. In this study, we reasonably designed and synthesized only 10 compounds to discover effective CDK4/6 protein degraders. Among the newly synthesized compounds, 7f achieved dual degradation of CDK4/6 protein, with DC50 values of 10.5 and 2.5 nM, respectively. Compound 7f also exhibited inhibitory proliferative activity against Jurkat cells with an IC50 value of 0.18 μM. Furthermore, 7f induced cell apoptosis and G1 phase cell cycle arrest in a dose-dependent manner in Jurkat cells. In conclusion, these findings demonstrate the potential of 7f as a CDK4/6 degrader and a potential therapeutic strategy against cancer, thereby expanding the potential of CDK4/6 dual PROTACs.

L-Ornithine derivatives with structural hetaryl and alkyl moiety: Synthesis and antibacterial activity

Objectives. Cationic amphiphiles and antimicrobial peptidomimetics are widely investigated as antibacterial agents due to their membrane-active mechanism of action. Particular attention is focused on the rational design of compounds in this class to achieve high antimicrobial activity. The aim of the present work is to synthesize bivalent cationic amphiphiles with L-ornithine as a branching element and evaluate the effectiveness of their antibacterial action. The compounds differ in terms of hydrophobicity due to the variation of N-terminal aliphatic amino acids in the polar block and alternation of dialkyl and alkyl-hetaryl radicals in the lipophilic block.Methods. For the synthesis of nonpolar fragments of amphiphiles, methods for the alkylation of amines with alkyl bromides in the presence of carbonate salts were used. The formation of amide bonds of L-ornithine derivatives with amino acids was carried out using the carbodiimide method. For the reaction products recovery from the reaction mixture, column chromatography on silica gel and aluminum oxide activated Brockmann Grade II was used. The antimicrobial activity of the synthesized compounds against gram-positive B. subtilis 534 and gram-negative E. coli M17 bacterial strains was evaluated. Minimum inhibitory concentration (MIC) values were recorded using a serial microdilution method in a nutrient medium.Results. Developed schemes for the preparation of bivalent cationic amphiphiles based on L-ornithine derivatives are presented. Differences in the structure of aliphatic amino acids (glycine, β-alanine, γ-aminobutyric acid (GABA)), in the length of alkyl radicals (C8, C12), or in the presence of an indole moiety, were used in the design of target compounds. The high antibacterial activity of the synthesized compounds was demonstrated. The most active compounds were lipoamino acids with terminal GABA residues and asymmetrical non-polar block (tryptamyl–dodecylamine). The MIC values were 0.39 μg/mL for gram-positive bacteria and 1.56 μg/mL for gram-negative bacteria. A GABA derivative with a symmetrical lipophilic moiety based on dioctylamine demonstrated activity with an MIC of 0.78 μg/mL against B. subtilis and 3.12 μg/mL against E. coli.Conclusions. Nine new lipoamino acid cationic bivalent amphiphiles based on L-ornithine were synthesized. The structure of the obtained compounds was confirmed by nuclear magnetic resonance 1H spectroscopy and mass spectrometry data. Leading compounds in antimicrobial activity against both gram-positive and gram-negative strains of bacteria were determined. The influence of the degree of lipophilicity in the asymmetric nonpolar block on the level of exhibited antimicrobial activity is demonstrated.

Functional screening and rational design of compounds targeting GPR132 to treat diabetes.

Chronic inflammation due to islet-residing macrophages plays key roles in the development of type 2 diabetes mellitus. By systematically profiling intra-islet lipid-transmembrane receptor signalling in islet-resident macrophages, we identified endogenous 9(S)-hydroxy-10,12-octadecadienoic acid-G-protein-coupled receptor 132 (GPR132)-Gi signalling as a significant contributor to islet macrophage reprogramming and found that GPR132 deficiency in macrophages reversed metabolic disorders in mice fed a high-fat diet. The cryo-electron microscopy structures of GPR132 bound with two endogenous agonists, N-palmitoylglycine and 9(S)-hydroxy-10,12-octadecadienoic acid, enabled us to rationally design both GPR132 agonists and antagonists with high potency and selectivity through stepwise translational approaches. We ultimately identified a selective GPR132 antagonist, NOX-6-18, that modulates macrophage reprogramming within pancreatic islets, decreases weight gain and enhances glucose metabolism in mice fed a high-fat diet. Our study not only illustrates that intra-islet lipid signalling contributes to islet macrophage reprogramming but also provides a broadly applicable strategy for the identification of important G-protein-coupled receptor targets in pathophysiological processes, followed by the rational design of therapeutic leads for refractory diseases such as diabetes.

Indol-3-ylglyoxylamide as Privileged Scaffold in Medicinal Chemistry.

In recent years, indolylglyoxylamide-based derivatives have received much attention due to their application in drug design and discovery, leading to the development of a wide array of compounds that have shown a variety of pharmacological activities. Combining the indole nucleus, already validated as a "privileged structure," with the glyoxylamide function allowed for an excellent template to be obtained that is suitable to a great number of structural modifications aimed at permitting interaction with specific molecular targets and producing desirable therapeutic effects. The present review provides insight into how medicinal chemists have elegantly exploited the indolylglyoxylamide moiety to obtain potentially useful drugs, with a particular focus on compounds exhibiting activity in in vivo models or reaching clinical trials. All in all, this information provides exciting new perspectives on existing data that can be useful in further design of indolylglyoxylamide-based molecules with interesting pharmacological profiles. The aim of this report is to present an update of collection data dealing with the employment of this moiety in the rational design of compounds that are able to interact with a specific target, referring to the last 20 years.

Design, synthesis, in silico, and in vitro evaluation of benzylbenzimidazolone derivatives as potential drugs on α-glucosidase and glucokinase as pharmacological targets.

Benzimidazolones have shown biological activities, including antihyperglycemic and hypoglycemic, by inhibiting or activating of α-glu and GK. The aim of this study is the rational design of compounds using in silico assays to delimitate the selection of structures to synthesize and the in vitro evaluation of benzimidazolone derivatives in blood glucose control. A docking of 23 benzimidazolone derivatives was performed; selecting the compounds with better in silico profiles to synthesize by microwave-irradiation/conventional heat and evaluate in enzymatic in vitro evaluation. Compounds 2k, 2m, 2r, and 2s presented the best in silico profiles, showing good affinity energy (-10.9 to -8.6 kcal mol-1) and binding with catalytic-amino acids. They were synthesized at 70 °C and 24 h using DMF as the solvent and potassium carbonate (yield: 22-38%). The results with α-glu showed moderate inhibition of 2k (14 ± 1.23-29 ± 0.45), 2m (12 ± 2.21-36 ± 0.30), 2r (7 ± 2.21-13 ± 1.34), and 2s (11 ± 0.74-35 ± 2.95) at evaluated concentrations (0.1 to 100 μg mL-1). The GK activation assay showed an enzymatic activity increase; compound 2k increased 1.31 and 2.83 more than normal activity, 2m (2.13-fold), 2s (2.86 and 3.74-fold) at 100 and 200 μg mL-1 respectively. The present study showed that the 2s derivative presents moderate potential as an α-glu inhibitor and a good activator potential of GK, suggesting that this compound is a good candidate for blood glucose control through antihyperglycemic and hypoglycemic mechanisms.

Property space mapping of Pseudomonas aeruginosa permeability to small molecules

Two membrane cell envelopes act as selective permeability barriers in Gram-negative bacteria, protecting cells against antibiotics and other small molecules. Significant efforts are being directed toward understanding how small molecules permeate these barriers. In this study, we developed an approach to analyze the permeation of compounds into Gram-negative bacteria and applied it to Pseudomonas aeruginosa, an important human pathogen notorious for resistance to multiple antibiotics. The approach uses mass spectrometric measurements of accumulation of a library of structurally diverse compounds in four isogenic strains of P. aeruginosa with varied permeability barriers. We further developed a machine learning algorithm that generates a deterministic classification model with minimal synonymity between the descriptors. This model predicted good permeators into P. aeruginosa with an accuracy of 89% and precision above 58%. The good permeators are broadly distributed in the property space and can be mapped to six distinct regions representing diverse chemical scaffolds. We posit that this approach can be used for more detailed mapping of the property space and for rational design of compounds with high Gram-negative permeability.

Structure‐Guided Design of G‐Protein‐Coupled Receptor Polypharmacology

AbstractMany diseases are polygenic and can only be treated efficiently with drugs that modulate multiple targets. However, rational design of compounds with multi‐target profiles is rarely pursued because it is considered too difficult, in particular if the drug must enter the central nervous system. Here, a structure‐based strategy to identify dual‐target ligands of G‐protein‐coupled receptors is presented. We use this approach to design compounds that both antagonize the A2A adenosine receptor and activate the D2 dopamine receptor, which have excellent potential as antiparkinson drugs. Atomic resolution models of the receptors guided generation of a chemical library with compounds designed to occupy orthosteric and secondary binding pockets in both targets. Structure‐based virtual screens identified ten compounds, of which three had affinity for both targets. One of these scaffolds was optimized to nanomolar dual‐target activity and showed the predicted pharmacodynamic effect in a rat model of Parkinsonism.

Structure-Guided Design of G-Protein-Coupled Receptor Polypharmacology.

Many diseases are polygenic and can only be treated efficiently with drugs that modulate multiple targets. However, rational design of compounds with multi‐target profiles is rarely pursued because it is considered too difficult, in particular if the drug must enter the central nervous system. Here, a structure‐based strategy to identify dual‐target ligands of G‐protein‐coupled receptors is presented. We use this approach to design compounds that both antagonize the A2A adenosine receptor and activate the D2 dopamine receptor, which have excellent potential as antiparkinson drugs. Atomic resolution models of the receptors guided generation of a chemical library with compounds designed to occupy orthosteric and secondary binding pockets in both targets. Structure‐based virtual screens identified ten compounds, of which three had affinity for both targets. One of these scaffolds was optimized to nanomolar dual‐target activity and showed the predicted pharmacodynamic effect in a rat model of Parkinsonism.

Conformational and Substituent Effects on The Rehybridization of Boron in β‐Substituted Ethylboranes

AbstractThe chemistry of boron‐containing molecules pervades the study of biomolecules, new materials, organic synthesis, and beyond. In addition, the stereochemistry and rehybridization capability of boron in molecules influence the reactivity and selectivity of reactions. Therefore, conformational analysis and the study of substituent effects on the modulation of the sp2–sp3 character of boron in β‐substituted ethylboranes may be valuable for the rational design of compounds in which the empty pB orbital plays a central role in the reaction intermediate. Accordingly, we have quantum‐chemically studied the orbital interactions and bonding properties that govern the conformational behavior and boron hybridization in β‐substituted ethylboranes (substituents, X=H, CH3, F, Cl, NH2, PH2, OH, and SH). Conformations with an anti arrangement along the B−C−C−X path are generally preferred over syn and/or gauche conformations, while a B−H bond tends to eclipse with the C−C bond rather than with a less sterically hindered C−H bond. However, the s character of boron relative to the C−B bond strongly decreases in the syn conformation, where X can participate as an electron donor to the empty pB orbital. Indeed, such interaction makes the syn conformation of 2‐aminoethylborane the single stable conformer in the gas phase.

Ca2+-dependent mechanism of membrane insertion and destabilization by the SARS-CoV-2 fusion peptide

Cell penetration after recognition of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) virus by the ACE2 receptor and the fusion of its viral envelope membrane with cellular membranes are the early steps of infectivity. A region of the Spike protein of the virus, identified as the “fusion peptide” (FP), is liberated at its N-terminal site by a specific cleavage occurring in concert with the interaction of the receptor-binding domain of the Spike. Studies have shown that penetration is enhanced by the required binding of Ca2+ ions to the FPs of coronaviruses, but the mechanisms of membrane insertion and destabilization remain unclear. We have predicted the preferred positions of Ca2+ binding to the SARS-CoV-2-FP, the role of Ca2+ ions in mediating peptide-membrane interactions, the preferred mode of insertion of the Ca2+-bound SARS-CoV-2-FP, and consequent effects on the lipid bilayer from extensive atomistic molecular dynamics simulations and trajectory analyses. In a systematic sampling of the interactions of the Ca2+-bound peptide models with lipid membranes, SARS-CoV-2-FP penetrated the bilayer and disrupted its organization only in two modes involving different structural domains. In one, the hydrophobic residues F833/I834 from the middle region of the peptide are inserted. In the other, more prevalent mode, the penetration involves residues L822/F823 from the LLF motif, which is conserved in CoV-2-like viruses, and is achieved by the binding of Ca2+ ions to the D830/D839 and E819/D820 residue pairs. FP penetration is shown to modify the molecular organization in specific areas of the bilayer, and the extent of membrane binding of the SARS-CoV-2 FP is significantly reduced in the absence of Ca2+ ions. These findings provide novel mechanistic insights regarding the role of Ca2+ in mediating SARS-CoV-2 fusion and provide a detailed structural platform to aid the ongoing efforts in rational design of compounds to inhibit SARS-CoV-2 cell entry.

Positive Allosteric Modulators of Metabotropic Glutamate Receptor 5 as Tool Compounds to Study Signaling Bias.

Positive allosteric modulation of metabotropic glutamate subtype 5 (mGlu5) receptor has emerged as a potential new therapeutic strategy for the treatment of schizophrenia and cognitive impairments. However, positive allosteric modulator (PAM) agonist activity has been associated with adverse side effects, and neurotoxicity has also been observed for pure PAMs. The structural and pharmacological basis of therapeutic versus adverse mGlu5 PAM in vivo effects remains unknown. Thus, gaining insights into the signaling fingerprints, as well as the binding kinetics of structurally diverse mGlu5 PAMs, may help in the rational design of compounds with desired properties. We assessed the binding and signaling profiles of N-methyl-5-(phenylethynyl)pyrimidin-2-amine (MPPA), 3-cyano-N-(2,5-diphenylpyrazol-3-yl)benzamide (CDPPB), and 1-[4-(4-chloro-2-fluoro-phenyl)piperazin-1-yl]-2-(4-pyridylmethoxy)ethenone [compound 2c, a close analog of 1-(4-(2-chloro-4-fluorophenyl)piperazin-1-yl)-2-(pyridin-4-ylmethoxy)ethanone] in human embryonic kidney 293A cells stably expressing mGlu5 using Ca2+ mobilization, inositol monophosphate (IP1) accumulation, extracellular signal-regulated kinase 1/2 (ERK1/2) phosphorylation, and receptor internalization assays. Of the three allosteric ligands, only CDPPB had intrinsic agonist efficacy, and it also had the longest receptor residence time and highest affinity. MPPA was a biased PAM, showing higher positive cooperativity with orthosteric agonists in ERK1/2 phosphorylation and Ca2+ mobilization over IP1 accumulation and receptor internalization. In primary cortical neurons, all three PAMs showed stronger positive cooperativity with (S)-3,5-dihydroxyphenylglycine (DHPG) in Ca2+ mobilization over IP1 accumulation. Our characterization of three structurally diverse mGlu5 PAMs provides further molecular pharmacological insights and presents the first assessment of PAM-mediated mGlu5 internalization. SIGNIFICANCE STATEMENT: Enhancing metabotropic glutamate receptor subtype 5 (mGlu5) activity is a promising strategy to treat cognitive and positive symptoms in schizophrenia. It is increasingly evident that positive allosteric modulators (PAMs) of mGlu5 are not all equal in preclinical models; there remains a need to better understand the molecular pharmacological properties of mGlu5 PAMs. This study reports detailed characterization of the binding and functional pharmacological properties of mGlu5 PAMs and is the first study of the effects of mGlu5 PAMs on receptor internalization.

Molecular Characterization of the β2-like Octopamine Receptor of Helicoverpa armigera.

Helicoverpa armigera is a devastating polyphagous and cosmopolitan crop pest. There are reports of this insect being resistant to a variety of pesticides raising concern worldwide. The Octopamine (OA) binding β2-like receptor (OAR), a GPCR, is widely distributed in the nervous system of the insect and plays essential roles in the physiology and development and thus is an important target for insecticides. Yet, the molecular characterization of the H. armigera OAR (HarmOAR) and rational design of compounds based on this receptor is lacking. As a first step, we performed multiple sequence alignment of all insect OARs, which revealed that the sequences contained all conserved class A GPCR motifs. Phylogenetic studies showed clade-specific variations in the protein sequences primarily arising owing to differences in the ICL3 loop region. Further, a structural model of HarmOAR was built using the inactive human β2AR as a template. 0.9µs atomistic simulations revealed conserved inter helical contacts and water molecules of HarmOAR. The detailed binding of octopamine was studied using molecular docking and 0.3µs atomistic simulations. Twenty-two insecticides active against octopamine receptors of other insects were compiled and docked to HarmOAR followed by rescoring with binding free energies to prioritize them for H. armigera. Our study suggests α-terpineol to be a good candidate as an insecticide or insect repellent for Helicoverpa armigera.

Exploring chemical compound space with quantum-based machine learning.

Rational design of compounds with specific properties requires understanding and fast evaluation of molecular properties throughout chemical compound space - the huge set of all potentially stable molecules. Recent advances in combining quantum-mechanical calculations with machine learning provide powerful tools for exploring wide swathes of chemical compound space. We present our perspective on this exciting and quickly developing field by discussing key advances in the development and applications of quantum-mechanics-based machine-learning methods to diverse compounds and properties, and outlining the challenges ahead. We argue that significant progress in the exploration and understanding of chemical compound space can be made through a systematic combination of rigorous physical theories, comprehensive synthetic data sets of microscopic and macroscopic properties, and modern machine-learning methods that account for physical and chemical knowledge.

Mechanism and site of action of big dynorphin on ASIC1a

Acid-sensing ion channels (ASICs) are proton-gated cation channels that contribute to neurotransmission, as well as initiation of pain and neuronal death following ischemic stroke. As such, there is a great interest in understanding the in vivo regulation of ASICs, especially by endogenous neuropeptides that potently modulate ASICs. The most potent endogenous ASIC modulator known to date is the opioid neuropeptide big dynorphin (BigDyn). BigDyn is up-regulated in chronic pain and increases ASIC-mediated neuronal death during acidosis. Understanding the mechanism and site of action of BigDyn on ASICs could thus enable the rational design of compounds potentially useful in the treatment of pain and ischemic stroke. To this end, we employ a combination of electrophysiology, voltage-clamp fluorometry, synthetic BigDyn analogs, and noncanonical amino acid-mediated photocrosslinking. We demonstrate that BigDyn binding results in an ASIC1a closed resting conformation that is distinct from open and desensitized states induced by protons. Using alanine-substituted BigDyn analogs, we find that the BigDyn modulation of ASIC1a is primarily mediated through electrostatic interactions of basic amino acids in the BigDyn N terminus. Furthermore, neutralizing acidic amino acids in the ASIC1a extracellular domain reduces BigDyn effects, suggesting a binding site at the acidic pocket. This is confirmed by photocrosslinking using the noncanonical amino acid azidophenylalanine. Overall, our data define the mechanism of how BigDyn modulates ASIC1a, identify the acidic pocket as the binding site for BigDyn, and thus highlight this cavity as an important site for the development of ASIC-targeting therapeutics.

In silico screening of drug-like molecules for the treatment of cardiovascular diseases on the basis of five-membered privileged heterocycles

Among various heterocyclic systems, the derivatives of five-membered heterocycles are of special interest. Most of the above mentioned heterocycles are treatred as so-called privileged structures in modern medicinal chemistry. In silico screening among five-membered heterocycles of molecules for the treatment of cardiovascular diseases is actual.
 The aim of the work was the search for synthetic drug-like molecules based on functionalized five-membered heterocycles and related heterocyclic systems as an element of the theoretical platform for rational design of compounds acting on the cardiovascular system, and prediction of their possible mechanism of action.
 The objects of the study were derivatives of uncondensed and condensed five-membered heterocycles. In the work, in silico approaches were applied using the programs: Hyper-Chem, PASS, AutoDock, PROTOX.
 Based on the previous studies, focused sub-libraries of small synthetic drug-like molecules based on functionalized five-membered heterocycles and related heterocyclic systems have been selected. The compounds were divided on 12 groups. The optimization of the compound structures, the drug-like parameters calculation were carried out. The activity prediction, the acute toxicity level and docking studies to probable bio-targets which are related with cardiovascular drug mechanism of action have been carried out. It was shown that thiazole and thiadiazole based compounds possessed the highest calculated affinity levels to selected bio-targets. This is consistent with PASS-based prediction data.
 Diverse functionalized derivatives of five-membered heterocycles (thiazole, thiazolidine, thiadiazole, pyrazole, thiophene, triazole) and related fused heterocycles have been grouped in focused sub-libraries of compounds. it has been established that thiazole and thiadiazole based compounds are promising objects for directed synthesis and further modification as potential cardiovascular agents based on the prediction of biological activity, the calculation of affinity to potent bio-targets, and the prediction of the drug-like features and acute toxicity level. The prognostic values of the parameters of the above mentioned groups of compounds may be used as the element of theoretical platform for the search and de novo design of potential drugs for the treatment of cardiovascular diseases.

Search of Allosteric Inhibitors and Associated Proteins of an AKT-like Kinase from Trypanosoma cruzi.

Proteins associated to the PI3K/AKT/mTOR signaling pathway are widely used targets for cancer treatment, and in recent years they have also been evaluated as putative targets in trypanosomatids parasites, such as Trypanosoma cruzi. Here, we performed a virtual screening approach to find candidates that can bind regions on or near the Pleckstrin homology domain of an AKT-like protein in T. cruzi. The compounds were also evaluated in vitro. The in silico and experimental results allowed us to identify a set of compounds that can potentially alter the intracellular signaling pathway through the AKT-like kinase of the parasite; among them, a derivative of the pyrazolopyridine nucleus with an IC50 of 14.25 ± 1.00 μM against amastigotes of T. cruzi. In addition, we built a protein–protein interaction network of T. cruzi to understand the role of the AKT-like protein in the parasite, and look for additional proteins that can be postulated as possible novel molecular targets for the rational design of compounds against T. cruzi.

A Massively Parallel Selection of Small Molecule-RNA Motif Binding Partners Informs Design of an Antiviral from Sequence

Many RNAs cause disease; however, RNA is rarely exploited as a small-molecule drug target. Our programmatic focus is to define privileged RNA motif small-molecule interactions to enable the rational design of compounds that modulate RNA biology starting from only sequence. We completed a massive, library-versus-library screen that probed over 50 million binding events between RNA motifs and small molecules. The resulting data provide a rich encyclopedia of small-molecule RNA recognition patterns, defining chemotypes and RNA motifs that confer selective, avid binding. The resulting interaction maps were mined against the entire viral genome of hepatitis C virus (HCV). A small molecule was identified that avidly bound RNA motifs present in the HCV 30 UTR and inhibited viral replication while having no effect on host cells. Collectively, this study represents the first whole-genome pattern recognition between small molecules and RNA folds.

Design of two-tail compounds with rotationally fixed benzenesulfonamide ring as inhibitors of carbonic anhydrases

Rational design of compounds that would bind specific pockets of the target proteins is a difficult task in drug design. The 12 isoforms of catalytically active human carbonic anhydrases (CAs) have highly similar active sites that make it difficult to design inhibitors selective for one or several CA isoforms. A series of CA inhibitors based on 2-chloro/bromo-benzenesulfonamide that is largely fixed in the CA active site together with one or two tails yielded compounds that were synthesized and evaluated as inhibitors of CA isoforms. Introduction of a second tail had significant influence on the binding affinity and two-tailed compounds in most cases provided high affinity and selectivity for CA IX and CA XIV. The contacts between several compounds and CA amino acids were determined by X-ray crystallography. Together with the intrinsic enthalpy and entropy of binding they provided the structure-thermodynamics correlations for this series of compounds with the insight how to rationally build compounds with desired CA isoform as a target.

Machine Learning Classification Models to Improve the Docking-based Screening: A Case of PI3K-Tankyrase Inhibitors.

One of the major challenges in the current drug discovery is the improvement of the docking-based virtual screening performance. It is especially important in the rational design of compounds with desired polypharmacology or selectivity profiles. To address this problem, we present a methodology for the development of target-specific scoring functions possessing high screening power. These scoring functions were built using the machine learning methods for the dual target inhibitors of PI3Kα and tankyrase, promising targets for colorectal cancer therapy. The Deep Neural Network models achieve the external test AUC ROC values of 0.96 and 0.93 for the random split and 0.90 and 0.84 for the time-based split of the PI3Kα and tankyrase inhibitors, respectively. In addition, the impact of the training set size and the actives/decoys ratio on the model quality was assessed. The study demonstrates that the optimized scoring functions could significantly improve the docking screening power for each individual target. This is very useful in the design of multitarget or selective drugs wherein the screening filters are applied in sequence.