Machine-learning approaches in drug discovery: methods and applications

Mapping the dark space of chemical reactions with extended nanomole synthesis and MALDI-TOF MS

Understanding the practical limitations of chemical reactions is critically important for efficiently planning the synthesis of compounds in pharmaceutical, agrochemical, and specialty chemical research and development. However, literature reports of the scope of new reactions are often cursory and biased toward successful results, severely limiting the ability to predict reaction outcomes for untested substrates. We herein illustrate strategies for carrying out large-scale surveys of chemical reactivity by using a material-sparing nanomole-scale automated synthesis platform with greatly expanded synthetic scope combined with ultrahigh-throughput matrix-assisted laser desorption/ionization-time-of-flight mass spectrometry (MALDI-TOF MS).

Sachem: a chemical cartridge for high-performance substructure search

BACKGROUND

Structure search is one of the valuable capabilities of small-molecule databases. Fingerprint-based screening methods are usually employed to enhance the search performance by reducing the number of calls to the verification procedure. In substructure search, fingerprints are designed to capture important structural aspects of the molecule to aid the decision about whether the molecule contains a given substructure. Currently available cartridges typically provide acceptable search performance for processing user queries, but do not scale satisfactorily with dataset size.

RESULTS

We present Sachem, a new open-source chemical cartridge that implements two substructure search methods: The first is a performance-oriented reimplementation of substructure indexing based on the OrChem fingerprint, and the second is a novel method that employs newly designed fingerprints stored in inverted indices. We assessed the performance of both methods on small, medium, and large datasets containing 1, 10, and 94 million compounds, respectively. Comparison of Sachem with other freely available cartridges revealed improvements in overall performance, scaling potential and screen-out efficiency.

CONCLUSIONS

The Sachem cartridge allows efficient substructure searches in databases of all sizes. The sublinear performance scaling of the second method and the ability to efficiently query large amounts of pre-extracted information may together open the door to new applications for substructure searches.

Recent advances for identification of new scaffolds and drug targets for Mycobacterium tuberculosis

Tuberculosis (TB) is a leading cause of mortality and morbidity with an estimated 1.7 billion people latently infected with the pathogen worldwide. Clinically, TB infection presents itself as an asymptomatic infection, which gradually manifests to life threatening disease. The emergence of various drug resistant strains of Mycobacterium tuberculosis and lengthy duration of chemotherapy are major challenges in the field of TB drug development. Hence, there is an urgent need to develop scaffolds that possess a novel mechanism of action, can shorten the duration of therapy, and are active against both drug resistant and susceptible strains. In this review, we will discuss recent progress made in the field of TB drug development with emphasis on screening methods and drug targets from M. tuberculosis. The current review provides insights into mechanism of action of new scaffolds that are being evaluated in various stages of clinical trials. © 2018 IUBMB Life, 70(9):905-916, 2018.

Identification of Estrogen Receptor α Antagonists from Natural Products via In Vitro and In Silico Approaches

Estrogen receptor α (ERα) is a successful target for ER-positive breast cancer and also reported to be relevant in many other diseases. Selective estrogen receptor modulators (SERMs) make a good therapeutic effect in clinic. Because of the drug resistance and side effects of current SERMs, the discovery of new SERMs is given more and more attention. Virtual screening is a validated method to high effectively to identify novel bioactive small molecules. Ligand-based machine learning methods and structure-based molecular docking were first performed for identification of ERα antagonist from in-house natural product library. Naive Bayesian and recursive partitioning models with two kinds of descriptors were built and validated based on training set, test set, and external test set and then were utilized for distinction of active and inactive compounds. Totally, 162 compounds were predicted as ER antagonists and were further evaluated by molecular docking. According to docking score, we selected 8 representative compounds for both ERα competitor assay and luciferase reporter gene assay. Genistein, daidzein, phloretin, ellagic acid, ursolic acid, (-)-epigallocatechin-3-gallate, kaempferol, and naringenin exhibited different levels for antagonistic activity against ERα. These studies validated the feasibility of machine learning methods for predicting bioactivities of ligands and provided better insight into the natural products acting as estrogen receptor modulator, which are important lead compounds for future new drug design.

BitterSweetForest: A Random Forest Based Binary Classifier to Predict Bitterness and Sweetness of Chemical Compounds

Taste of a chemical compound present in food stimulates us to take in nutrients and avoid poisons. However, the perception of taste greatly depends on the genetic as well as evolutionary perspectives. The aim of this work was the development and validation of a machine learning model based on molecular fingerprints to discriminate between sweet and bitter taste of molecules. BitterSweetForest is the first open access model based on KNIME workflow that provides platform for prediction of bitter and sweet taste of chemical compounds using molecular fingerprints and Random Forest based classifier. The constructed model yielded an accuracy of 95% and an AUC of 0.98 in cross-validation. In independent test set, BitterSweetForest achieved an accuracy of 96% and an AUC of 0.98 for bitter and sweet taste prediction. The constructed model was further applied to predict the bitter and sweet taste of natural compounds, approved drugs as well as on an acute toxicity compound data set. BitterSweetForest suggests 70% of the natural product space, as bitter and 10% of the natural product space as sweet with confidence score of 0.60 and above. 77% of the approved drug set was predicted as bitter and 2% as sweet with a confidence score of 0.75 and above. Similarly, 75% of the total compounds from acute oral toxicity class were predicted only as bitter with a minimum confidence score of 0.75, revealing toxic compounds are mostly bitter. Furthermore, we applied a Bayesian based feature analysis method to discriminate the most occurring chemical features between sweet and bitter compounds using the feature space of a circular fingerprint.

Transfer and Multi-task Learning in QSAR Modeling: Advances and Challenges

Medicinal chemistry projects involve some steps aiming to develop a new drug, such as the analysis of biological targets related to a given disease, the discovery and the development of drug candidates for these targets, performing parallel biological tests to validate the drug effectiveness and side effects. Approaches as quantitative study of activity-structure relationships (QSAR) involve the construction of predictive models that relate a set of descriptors of a chemical compound series and its biological activities with respect to one or more targets in the human body. Datasets used to perform QSAR analyses are generally characterized by a small number of samples and this makes them more complex to build accurate predictive models. In this context, transfer and multi-task learning techniques are very suitable since they take information from other QSAR models to the same biological target, reducing efforts and costs for generating new chemical compounds. Therefore, this review will present the main features of transfer and multi-task learning studies, as well as some applications and its potentiality in drug design projects.

The benefits of in silico modeling to identify possible small-molecule drugs and their off-target interactions

The research into the use of small molecules as drugs continues to be a key driver in the development of molecular databases, computer-aided drug design software and collaborative platforms. The evolution of computational approaches is driven by the essential criteria that a drug molecule has to fulfill, from the affinity to targets to minimal side effects while having adequate absorption, distribution, metabolism, and excretion (ADME) properties. A combination of ligand- and structure-based drug development approaches is already used to obtain consensus predictions of small molecule activities and their off-target interactions. Further integration of these methods into easy-to-use workflows informed by systems biology could realize the full potential of available data in the drug discovery and reduce the attrition of drug candidates.

Similarity search combined with docking and molecular dynamics for novel hAChE inhibitor scaffolds

The main purpose of this study was to address the performance of virtual screening methods based on ligands and the protein structure of acetylcholinesterase (AChE) in order to retrieve novel human AChE (hAChE) inhibitors. In addition, a protocol was developed to identify novel hit compounds and propose new promising AChE inhibitors from the ZINC database with 10 million commercially available compounds. In this sense, 3D similarity searches using rapid overlay of chemical structures and similarity analysis through comparison of electrostatic overlay of docked hits were used to retrieve AChE inhibitors from collected databases. Molecular dynamics simulation of 100 ns was carried out to study the best docked compounds from similarity searches. Some key residues were identified as crucial for the dual binding mode of inhibitor with the interaction site. All results indicated the relevant use of EON and docking strategy for identifying novel hit compounds as promising potential anticholinesterase candidates, and seven new structures were selected as potential hAChE inhibitors. Graphical abstract Compound N01 in the 4M0E hAChE crystallography structure from docking results. Yellow dashed lines Hydrogen bonds, blue dashed lines π-stacking interactions, green dashed lines cation-π interactions.

Ensemble learning method for the prediction of new bioactive molecules

Pharmacologically active molecules can provide remedies for a range of different illnesses and infections. Therefore, the search for such bioactive molecules has been an enduring mission. As such, there is a need to employ a more suitable, reliable, and robust classification method for enhancing the prediction of the existence of new bioactive molecules. In this paper, we adopt a recently developed combination of different boosting methods (Adaboost) for the prediction of new bioactive molecules. We conducted the research experiments utilizing the widely used MDL Drug Data Report (MDDR) database. The proposed boosting method generated better results than other machine learning methods. This finding suggests that the method is suitable for inclusion among the in silico tools for use in cheminformatics, computational chemistry and molecular biology.

Computer-Aided Drug Design: An Overview

The term drug design describes the search of novel compounds with biological activity, on a systematic basis. In its most common form, it involves modification of a known active scaffold or linking known active scaffolds, although de novo drug design (i.e., from scratch) is also possible. Though highly interrelated, identification of active scaffolds should be conceptually separated from drug design. Traditionally, the drug design process has focused on the molecular determinants of the interactions between the drug and its known or intended molecular target. Nevertheless, current drug design also takes into consideration other relevant processes than influence drug efficacy and safety (e.g., bioavailability, metabolic stability, interaction with antitargets).This chapter provides an overview on possible approaches to identify active scaffolds (including in silico approximations to approach that task) and computational methods to guide the subsequent optimization process. It also discusses in which situations each of the overviewed techniques is more appropriate.

Utilizing random Forest QSAR models with optimized parameters for target identification and its application to target-fishing server

Background

The identification of target molecules is important for understanding the mechanism of “target deconvolution” in phenotypic screening and “polypharmacology” of drugs. Because conventional methods of identifying targets require time and cost, in-silico target identification has been considered an alternative solution. One of the well-known in-silico methods of identifying targets involves structure activity relationships (SARs). SARs have advantages such as low computational cost and high feasibility; however, the data dependency in the SAR approach causes imbalance of active data and ambiguity of inactive data throughout targets.

Results

We developed a ligand-based virtual screening model comprising 1121 target SAR models built using a random forest algorithm. The performance of each target model was tested by employing the ROC curve and the mean score using an internal five-fold cross validation. Moreover, recall rates for top-k targets were calculated to assess the performance of target ranking. A benchmark model using an optimized sampling method and parameters was examined via external validation set. The result shows recall rates of 67.6% and 73.9% for top-11 (1% of the total targets) and top-33, respectively. We provide a website for users to search the top-k targets for query ligands available publicly at http://rfqsar.kaist.ac.kr.

Conclusions

The target models that we built can be used for both predicting the activity of ligands toward each target and ranking candidate targets for a query ligand using a unified scoring scheme. The scores are additionally fitted to the probability so that users can estimate how likely a ligand–target interaction is active. The user interface of our web site is user friendly and intuitive, offering useful information and cross references.

Electronic supplementary material

The online version of this article (10.1186/s12859-017-1960-x) contains supplementary material, which is available to authorized users.

Can human experts predict solubility better than computers?

In this study, we design and carry out a survey, asking human experts to predict the aqueous solubility of druglike organic compounds. We investigate whether these experts, drawn largely from the pharmaceutical industry and academia, can match or exceed the predictive power of algorithms. Alongside this, we implement 10 typical machine learning algorithms on the same dataset. The best algorithm, a variety of neural network known as a multi-layer perceptron, gave an RMSE of 0.985 log S units and an R² of 0.706. We would not have predicted the relative success of this particular algorithm in advance. We found that the best individual human predictor generated an almost identical prediction quality with an RMSE of 0.942 log S units and an R² of 0.723. The collection of algorithms contained a higher proportion of reasonably good predictors, nine out of ten compared with around half of the humans. We found that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median generated excellent predictivity. While our consensus human predictor achieved very slightly better headline figures on various statistical measures, the difference between it and the consensus machine learning predictor was both small and statistically insignificant. We conclude that human experts can predict the aqueous solubility of druglike molecules essentially equally well as machine learning algorithms. We find that, for either humans or algorithms, combining individual predictions into a consensus predictor by taking their median is a powerful way of benefitting from the wisdom of crowds.

In silico ligand-based modeling of hBACE-1 inhibitors

Alzheimer's disease is a chronic neurodegenerative disease affecting more than 30 million people worldwide. Development of small molecule inhibitors of human β-secretase 1 (hBACE-1) is being the focus of pharmaceutical industry for the past 15-20 years. Here, we successfully applied multiple ligand-based in silico modeling techniques to understand the inhibitory activities of a diverse set of small molecule hBACE-1 inhibitors reported in the scientific literature. Strikingly, the use of only a small subset of 230 (13%) molecules allowed us to develop quality models that performed reasonably well on the validation set of 1,476 (87%) inhibitors. Varying the descriptor sets and the complexity of the modeling techniques resulted in only minor improvements to the model's performance. The current results demonstrate that predictive models can be built by choosing appropriate modeling techniques in spite of using small datasets consisting of diverse chemical classes, a scenario typical in triaging of high-throughput screening results to identify false negatives. We hope that these encouraging results will help the community to develop more predictive models that would support research efforts for the debilitating Alzheimer's disease. Additionally, the integrated diversity of the techniques employed will stimulate scientists in the field to use in silico statistical modeling techniques like these to derive better models to help advance the drug discovery projects faster.

A Systematic Review of Computational Drug Discovery, Development, and Repurposing for Ebola Virus Disease Treatment

Ebola virus disease (EVD) is a deadly global public health threat, with no currently approved treatments. Traditional drug discovery and development is too expensive and inefficient to react quickly to the threat. We review published research studies that utilize computational approaches to find or develop drugs that target the Ebola virus and synthesize its results. A variety of hypothesized and/or novel treatments are reported to have potential anti-Ebola activity. Approaches that utilize multi-targeting/polypharmacology have the most promise in treating EVD.

Discovery of Key Physicochemical, Structural, and Spatial Properties of RNA-Targeted Bioactive Ligands

While a myriad non-coding RNAs are known to be essential in cellular processes and misregulated in diseases, the development of RNA-targeted small molecule probes has met with limited success. To elucidate the guiding principles for selective small molecule/RNA recognition, we analyzed cheminformatic and shape-based descriptors for 104 RNA-targeted ligands with demonstrated biological activity (RNA-targeted BIoactive ligaNd Database, R-BIND). We then compared R-BIND to both FDA-approved small molecule drugs and RNA ligands without reported bioactivity. Several striking trends emerged for bioactive RNA ligands, including: 1) Compliance to medicinal chemistry rules, 2) distinctive structural features, and 3) enrichment in rod-like shapes over others. This work provides unique insights that directly facilitate the selection and synthesis of RNA-targeted libraries with the goal of efficiently identifying selective small molecule ligands for therapeutically relevant RNAs.

Decrypting Strong and Weak Single-Walled Carbon Nanotubes Interactions with Mitochondrial Voltage-Dependent Anion Channels Using Molecular Docking and Perturbation Theory

The current molecular docking study provided the Free Energy of Binding (FEB) for the interaction (nanotoxicity) between VDAC mitochondrial channels of three species (VDAC1-Mus musculus, VDAC1-Homo sapiens, VDAC2-Danio rerio) with SWCNT-H, SWCNT-OH, SWCNT-COOH carbon nanotubes. The general results showed that the FEB values were statistically more negative (p < 0.05) in the following order: (SWCNT-VDAC2-Danio rerio) > (SWCNT-VDAC1-Mus musculus) > (SWCNT-VDAC1-Homo sapiens) > (ATP-VDAC). More negative FEB values for SWCNT-COOH and OH were found in VDAC2-Danio rerio when compared with VDAC1-Mus musculus and VDAC1-Homo sapiens (p < 0.05). In addition, a significant correlation (0.66 > r² > 0.97) was observed between n-Hamada index and VDAC nanotoxicity (or FEB) for the zigzag topologies of SWCNT-COOH and SWCNT-OH. Predictive Nanoparticles-Quantitative-Structure Binding-Relationship models (nano-QSBR) for strong and weak SWCNT-VDAC docking interactions were performed using Perturbation Theory, regression and classification models. Thus, 405 SWCNT-VDAC interactions were predicted using a nano-PT-QSBR classifications model with high accuracy, specificity, and sensitivity (73–98%) in training and validation series, and a maximum AUROC value of 0.978. In addition, the best regression model was obtained with Random Forest (R² of 0.833, RMSE of 0.0844), suggesting an excellent potential to predict SWCNT-VDAC channel nanotoxicity. All study data are available at https://doi.org/10.6084/m9.figshare.4802320.v2.

VB-MK-LMF: fusion of drugs, targets and interactions using variational Bayesian multiple kernel logistic matrix factorization

BACKGROUND

Computational fusion approaches to drug-target interaction (DTI) prediction, capable of utilizing multiple sources of background knowledge, were reported to achieve superior predictive performance in multiple studies. Other studies showed that specificities of the DTI task, such as weighting the observations and focusing the side information are also vital for reaching top performance.

METHOD

We present Variational Bayesian Multiple Kernel Logistic Matrix Factorization (VB-MK-LMF), which unifies the advantages of (1) multiple kernel learning, (2) weighted observations, (3) graph Laplacian regularization, and (4) explicit modeling of probabilities of binary drug-target interactions.

RESULTS

VB-MK-LMF achieves significantly better predictive performance in standard benchmarks compared to state-of-the-art methods, which can be traced back to multiple factors. The systematic evaluation of the effect of multiple kernels confirm their benefits, but also highlights the limitations of linear kernel combinations, already recognized in other fields. The analysis of the effect of prior kernels using varying sample sizes sheds light on the balance of data and knowledge in DTI tasks and on the rate at which the effect of priors vanishes. This also shows the existence of "small sample size" regions where using side information offers significant gains. Alongside favorable predictive performance, a notable property of MF methods is that they provide a unified space for drugs and targets using latent representations. Compared to earlier studies, the dimensionality of this space proved to be surprisingly low, which makes the latent representations constructed by VB-ML-LMF especially well-suited for visual analytics. The probabilistic nature of the predictions allows the calculation of the expected values of hits in functionally relevant sets, which we demonstrate by predicting drug promiscuity. The variational Bayesian approximation is also implemented for general purpose graphics processing units yielding significantly improved computational time.

CONCLUSION

In standard benchmarks, VB-MK-LMF shows significantly improved predictive performance in a wide range of settings. Beyond these benchmarks, another contribution of our work is highlighting and providing estimates for further pharmaceutically relevant quantities, such as promiscuity, druggability and total number of interactions.

Optimization to the Phellinus experimental environment based on classification forecasting method

Phellinus is a kind of fungus and known as one of the elemental components in drugs to avoid cancer. With the purpose of finding optimized culture conditions for Phellinus production in the lab, plenty of experiments focusing on single factor were operated and large scale of experimental data was generated. In previous work, we used regression analysis and GA Gene-set based Genetic Algorithm (GA) to predict the production, but the data we used depended on experimental experience and only little part of the data was used. In this work we use the values of parameters involved in culture conditions, including inoculum size, PH value, initial liquid volume, temperature, seed age, fermentation time and rotation speed, to establish a high yield and a low yield classification model. Subsequently, a prediction model of BP neural network is established for high yield data set. GA is used to find the best culture conditions. The forecast accuracy rate more than 90% and the yield we got have a slight increase than the real yield.

From machine learning to deep learning: progress in machine intelligence for rational drug discovery

Machine intelligence, which is normally presented as artificial intelligence, refers to the intelligence exhibited by computers. In the history of rational drug discovery, various machine intelligence approaches have been applied to guide traditional experiments, which are expensive and time-consuming. Over the past several decades, machine-learning tools, such as quantitative structure-activity relationship (QSAR) modeling, were developed that can identify potential biological active molecules from millions of candidate compounds quickly and cheaply. However, when drug discovery moved into the era of 'big' data, machine learning approaches evolved into deep learning approaches, which are a more powerful and efficient way to deal with the massive amounts of data generated from modern drug discovery approaches. Here, we summarize the history of machine learning and provide insight into recently developed deep learning approaches and their applications in rational drug discovery. We suggest that this evolution of machine intelligence now provides a guide for early-stage drug design and discovery in the current big data era.

Metabolomics and Integrative Omics for the Development of Thai Traditional Medicine

In recent years, interest in studies of traditional medicine in Asian and African countries has gradually increased due to its potential to complement modern medicine. In this review, we provide an overview of Thai traditional medicine (TTM) current development, and ongoing research activities of TTM related to metabolomics. This review will also focus on three important elements of systems biology analysis of TTM including analytical techniques, statistical approaches and bioinformatics tools for handling and analyzing untargeted metabolomics data. The main objective of this data analysis is to gain a comprehensive understanding of the system wide effects that TTM has on individuals. Furthermore, potential applications of metabolomics and systems medicine in TTM will also be discussed.

Development of Theoretical Descriptors for Cytotoxicity Evaluation of Metallic Nanoparticles

Motivated by the recent development of quantitative structure-activity relationship (QSAR) methods in the area of nanotoxicology, we proposed an approach to develop additional descriptors based on results of first-principles calculations. For the evaluation of the biochemical activity of metallic nanoparticles, we consider two processes: ion extraction from the surface of a specimen to aqueous media and water dissociation on the surface. We performed calculations for a set of metals (Al, Fe, Cu, Ag, Au, and Pt). Taking into account the diversity of atomic structures of real metallic nanoparticles, we performed calculations for different models such as (001) and (111) surfaces, nanorods, and two different cubic nanoparticles of 0.6 and 0.3 nm size. Significant energy dependence of the processes from the selected model of nanoparticle suggests that for the correct description we should combine the calculations for several representative models. In addition to the descriptors of chemical activity of the metallic nanoparticles for the two studied processes, we propose descriptors for taking into account the dependence of chemical activity from the size and shape of nanoparticles. Routes to minimization of computational costs for these calculations are also discussed.

Classification of nervous system withdrawn and approved drugs with ToxPrint features via machine learning strategies

BACKGROUND AND OBJECTIVES

Early-phase virtual screening of candidate drug molecules plays a key role in pharmaceutical industry from data mining and machine learning to prevent adverse effects of the drugs. Computational classification methods can distinguish approved drugs from withdrawn ones. We focused on 6 data sets including maximum 110 approved and 110 withdrawn drugs for all and nervous system diseases to distinguish approved drugs from withdrawn ones.

METHODS

In this study, we used support vector machines (SVMs) and ensemble methods (EMs) such as boosted and bagged trees to classify drugs into approved and withdrawn categories. Also, we used CORINA Symphony program to identify Toxprint chemotypes including over 700 predefined chemotypes for determination of risk and safety assesment of candidate drug molecules. In addition, we studied nervous system withdrawn drugs to determine the key fragments with The ParMol package including gSpan algorithm.

RESULTS

According to our results, the descriptors named as the number of total chemotypes and bond CN_amine_aliphatic_generic were more significant descriptors. The developed Medium Gaussian SVM model reached 78% prediction accuracy on test set for drug data set including all disease. Here, bagged tree and linear SVM models showed 89% of accuracies for phycholeptics and psychoanaleptics drugs. A set of discriminative fragments in nervous system withdrawn drug (NSWD) data sets was obtained. These fragments responsible for the drugs removed from market were benzene, toluene, N,N-dimethylethylamine, crotylamine, 5-methyl-2,4-heptadiene, octatriene and carbonyl group.

CONCLUSION

This paper covers the development of computational classification methods to distinguish approved drugs from withdrawn ones. In addition, the results of this study indicated the identification of discriminative fragments is of significance to design a new nervous system approved drugs with interpretation of the structures of the NSWDs.

Medical big data: promise and challenges

The concept of big data, commonly characterized by volume, variety, velocity, and veracity, goes far beyond the data type and includes the aspects of data analysis, such as hypothesis-generating, rather than hypothesis-testing. Big data focuses on temporal stability of the association, rather than on causal relationship and underlying probability distribution assumptions are frequently not required. Medical big data as material to be analyzed has various features that are not only distinct from big data of other disciplines, but also distinct from traditional clinical epidemiology. Big data technology has many areas of application in healthcare, such as predictive modeling and clinical decision support, disease or safety surveillance, public health, and research. Big data analytics frequently exploits analytic methods developed in data mining, including classification, clustering, and regression. Medical big data analyses are complicated by many technical issues, such as missing values, curse of dimensionality, and bias control, and share the inherent limitations of observation study, namely the inability to test causality resulting from residual confounding and reverse causation. Recently, propensity score analysis and instrumental variable analysis have been introduced to overcome these limitations, and they have accomplished a great deal. Many challenges, such as the absence of evidence of practical benefits of big data, methodological issues including legal and ethical issues, and clinical integration and utility issues, must be overcome to realize the promise of medical big data as the fuel of a continuous learning healthcare system that will improve patient outcome and reduce waste in areas including nephrology.

Quantitative structure-activity relationship analysis and virtual screening studies for identifying HDAC2 inhibitors from known HDAC bioactive chemical libraries

Histone deacetylases (HDAC) are emerging as promising targets in cancer, neuronal diseases and immune disorders. Computational modelling approaches have been widely applied for the virtual screening and rational design of novel HDAC inhibitors. In this study, different machine learning (ML) techniques were applied for the development of models that accurately discriminate HDAC2 inhibitors form non-inhibitors. The obtained models showed encouraging results, with the global accuracy in the external set ranging from 0.83 to 0.90. Various aspects related to the comparison of modelling techniques, applicability domain and descriptor interpretations were discussed. Finally, consensus predictions of these models were used for screening HDAC2 inhibitors from four chemical libraries whose bioactivities against HDAC1, HDAC3, HDAC6 and HDAC8 have been known. According to the results of virtual screening assays, structures of some hits with pair-isoform-selective activity (between HDAC2 and other HDACs) were revealed. This study illustrates the power of ML-based QSAR approaches for the screening and discovery of potent, isoform-selective HDACIs.

Quantum probability ranking principle for ligand-based virtual screening

Chemical libraries contain thousands of compounds that need screening, which increases the need for computational methods that can rank or prioritize compounds. The tools of virtual screening are widely exploited to enhance the cost effectiveness of lead drug discovery programs by ranking chemical compounds databases in decreasing probability of biological activity based upon probability ranking principle (PRP). In this paper, we developed a novel ranking approach for molecular compounds inspired by quantum mechanics, called quantum probability ranking principle (QPRP). The QPRP ranking criteria would make an attempt to draw an analogy between the physical experiment and molecular structure ranking process for 2D fingerprints in ligand based virtual screening (LBVS). The development of QPRP criteria in LBVS has employed the concepts of quantum at three different levels, firstly at representation level, this model makes an effort to develop a new framework of molecular representation by connecting the molecular compounds with mathematical quantum space. Secondly, estimate the similarity between chemical libraries and references based on quantum-based similarity searching method. Finally, rank the molecules using QPRP approach. Simulated virtual screening experiments with MDL drug data report (MDDR) data sets showed that QPRP outperformed the classical ranking principle (PRP) for molecular chemical compounds.

Predicting DPP-IV inhibitors with machine learning approaches

Dipeptidyl peptidase IV (DPP-IV) is a promising Type 2 diabetes mellitus (T2DM) drug target. DPP-IV inhibitors prolong the action of glucagon-like peptide-1 (GLP-1) and gastric inhibitory peptide (GIP), improve glucose homeostasis without weight gain, edema, and hypoglycemia. However, the marketed DPP-IV inhibitors have adverse effects such as nasopharyngitis, headache, nausea, hypersensitivity, skin reactions and pancreatitis. Therefore, it is still expected for novel DPP-IV inhibitors with minimal adverse effects. The scaffolds of existing DPP-IV inhibitors are structurally diversified. This makes it difficult to build virtual screening models based upon the known DPP-IV inhibitor libraries using conventional QSAR approaches. In this paper, we report a new strategy to predict DPP-IV inhibitors with machine learning approaches involving naïve Bayesian (NB) and recursive partitioning (RP) methods. We built 247 machine learning models based on 1307 known DPP-IV inhibitors with optimized molecular properties and topological fingerprints as descriptors. The overall predictive accuracies of the optimized models were greater than 80%. An external test set, composed of 65 recently reported compounds, was employed to validate the optimized models. The results demonstrated that both NB and RP models have a good predictive ability based on different combinations of descriptors. Twenty "good" and twenty "bad" structural fragments for DPP-IV inhibitors can also be derived from these models for inspiring the new DPP-IV inhibitor scaffold design.

Improving scoring-docking-screening powers of protein-ligand scoring functions using random forest

The development of new protein-ligand scoring functions using machine learning algorithms, such as random forest, has been of significant interest. By efficiently utilizing expanded feature sets and a large set of experimental data, random forest based scoring functions (RFbScore) can achieve better correlations to experimental protein-ligand binding data with known crystal structures; however, more extensive tests indicate that such enhancement in scoring power comes with significant under-performance in docking and screening power tests compared to traditional scoring functions. In this work, to improve scoring-docking-screening powers of protein-ligand docking functions simultaneously, we have introduced a Δvina RF parameterization and feature selection framework based on random forest. Our developed scoring function Δvina RF20 , which employs 20 descriptors in addition to the AutoDock Vina score, can achieve superior performance in all power tests of both CASF-2013 and CASF-2007 benchmarks compared to classical scoring functions. The Δvina RF20 scoring function and its code are freely available on the web at: https://www.nyu.edu/projects/yzhang/DeltaVina. © 2016 Wiley Periodicals, Inc.

Computational methods in drug discovery

The process for drug discovery and development is challenging, time consuming and expensive. Computer-aided drug discovery (CADD) tools can act as a virtual shortcut, assisting in the expedition of this long process and potentially reducing the cost of research and development. Today CADD has become an effective and indispensable tool in therapeutic development. The human genome project has made available a substantial amount of sequence data that can be used in various drug discovery projects. Additionally, increasing knowledge of biological structures, as well as increasing computer power have made it possible to use computational methods effectively in various phases of the drug discovery and development pipeline. The importance of in silico tools is greater than ever before and has advanced pharmaceutical research. Here we present an overview of computational methods used in different facets of drug discovery and highlight some of the recent successes. In this review, both structure-based and ligand-based drug discovery methods are discussed. Advances in virtual high-throughput screening, protein structure prediction methods, protein-ligand docking, pharmacophore modeling and QSAR techniques are reviewed.

Global vision of druggability issues: applications and perspectives

During the preliminary stage of a drug discovery project, the lack of druggability information and poor target selection are the main causes of frequent failures. Elaborating on accurate computational druggability prediction methods is a requirement for prioritizing target selection, designing new drugs and avoiding side effects. In this review, we describe a survey of recently reported druggability prediction methods mainly based on networks, statistical pocket druggability predictions and virtual screening. An application for a frequent mutation of p53 tumor suppressor is presented, illustrating the complementarity of druggability prediction approaches, the remaining challenges and potential new drug development perspectives.

Use of machine learning approaches for novel drug discovery

INTRODUCTION

The use of computational tools in the early stages of drug development has increased in recent decades. Machine learning (ML) approaches have been of special interest, since they can be applied in several steps of the drug discovery methodology, such as prediction of target structure, prediction of biological activity of new ligands through model construction, discovery or optimization of hits, and construction of models that predict the pharmacokinetic and toxicological (ADMET) profile of compounds.

AREAS COVERED

This article presents an overview on some applications of ML techniques in drug design. These techniques can be employed in ligand-based drug design (LBDD) and structure-based drug design (SBDD) studies, such as similarity searches, construction of classification and/or prediction models of biological activity, prediction of secondary structures and binding sites docking and virtual screening.

EXPERT OPINION

Successful cases have been reported in the literature, demonstrating the efficiency of ML techniques combined with traditional approaches to study medicinal chemistry problems. Some ML techniques used in drug design are: support vector machine, random forest, decision trees and artificial neural networks. Currently, an important application of ML techniques is related to the calculation of scoring functions used in docking and virtual screening assays from a consensus, combining traditional and ML techniques in order to improve the prediction of binding sites and docking solutions.

Computational methods for prediction of in vitro effects of new chemical structures

BACKGROUND

With a constant increase in the number of new chemicals synthesized every year, it becomes important to employ the most reliable and fast in silico screening methods to predict their safety and activity profiles. In recent years, in silico prediction methods received great attention in an attempt to reduce animal experiments for the evaluation of various toxicological endpoints, complementing the theme of replace, reduce and refine. Various computational approaches have been proposed for the prediction of compound toxicity ranging from quantitative structure activity relationship modeling to molecular similarity-based methods and machine learning. Within the "Toxicology in the 21st Century" screening initiative, a crowd-sourcing platform was established for the development and validation of computational models to predict the interference of chemical compounds with nuclear receptor and stress response pathways based on a training set containing more than 10,000 compounds tested in high-throughput screening assays.

RESULTS

Here, we present the results of various molecular similarity-based and machine-learning based methods over an independent evaluation set containing 647 compounds as provided by the Tox21 Data Challenge 2014. It was observed that the Random Forest approach based on MACCS molecular fingerprints and a subset of 13 molecular descriptors selected based on statistical and literature analysis performed best in terms of the area under the receiver operating characteristic curve values. Further, we compared the individual and combined performance of different methods. In retrospect, we also discuss the reasons behind the superior performance of an ensemble approach, combining a similarity search method with the Random Forest algorithm, compared to individual methods while explaining the intrinsic limitations of the latter.

CONCLUSIONS

Our results suggest that, although prediction methods were optimized individually for each modelled target, an ensemble of similarity and machine-learning approaches provides promising performance indicating its broad applicability in toxicity prediction.

Ligand-based chemoinformatic discovery of a novel small molecule inhibitor targeting CDC25 dual specificity phosphatases and displaying in vitro efficacy against melanoma cells

CDC25 phosphatases are important regulators of the cell cycle and represent promising targets for anticancer drug discovery. We recently identified NSC 119915 as a new quinonoid CDC25 inhibitor with potent anticancer activity. In order to discover more active analogs of NSC 119915, we performed a range of ligand-based chemoinformatic methods against the full ZINC drug-like subset and the NCI lead-like set. Nine compounds (3, 5-9, 21, 24, and 25) were identified with Ki values for CDC25A, -B and -C ranging from 0.01 to 4.4 μM. One of these analogs, 7, showed a high antiproliferative effect on human melanoma cell lines, A2058 and SAN. Compound 7 arrested melanoma cells in G2/M, causing a reduction of the protein levels of CDC25A and, more consistently, of CDC25C. Furthermore, an intrinsic apoptotic pathway was induced, which was mediated by ROS, because it was reverted in the presence of antioxidant N-acetyl-cysteine (NAC). Finally, 7 decreased the protein levels of phosphorylated Akt and increased those of p53, thus contributing to the regulation of chemosensitivity through the control of downstream Akt pathways in melanoma cells. Taken together, our data emphasize that CDC25 could be considered as a possible oncotarget in melanoma cells and that compound 7 is a small molecule CDC25 inhibitor that merits to be further evaluated as a chemotherapeutic agent for melanoma, likely in combination with other therapeutic compounds.

Discovery of Novel Antischistosomal Agents by Molecular Modeling Approaches

Schistosomiasis, a chronic neglected tropical disease caused by Schistosoma worms, is reported in nearly 80 countries. Although the disease affects approximately 260 million people, the treatment relies exclusively on praziquantel, a drug discovered in the mid-1970s that lacks efficacy against the larval stages of the parasite. In addition, the dependence on a single treatment has raised concerns about drug resistance, and reduced susceptibility has already been found in laboratory and field isolates. Therefore, novel therapies for schistosomiasis are needed, and several approaches have been used to that end. One of these strategies, molecular modeling, has been increasingly integrated with experimental techniques, resulting in the discovery of novel antischistosomal agents.

Deep Artificial Neural Networks and Neuromorphic Chips for Big Data Analysis: Pharmaceutical and Bioinformatics Applications

Over the past decade, Deep Artificial Neural Networks (DNNs) have become the state-of-the-art algorithms in Machine Learning (ML), speech recognition, computer vision, natural language processing and many other tasks. This was made possible by the advancement in Big Data, Deep Learning (DL) and drastically increased chip processing abilities, especially general-purpose graphical processing units (GPGPUs). All this has created a growing interest in making the most of the potential offered by DNNs in almost every field. An overview of the main architectures of DNNs, and their usefulness in Pharmacology and Bioinformatics are presented in this work. The featured applications are: drug design, virtual screening (VS), Quantitative Structure-Activity Relationship (QSAR) research, protein structure prediction and genomics (and other omics) data mining. The future need of neuromorphic hardware for DNNs is also discussed, and the two most advanced chips are reviewed: IBM TrueNorth and SpiNNaker. In addition, this review points out the importance of considering not only neurons, as DNNs and neuromorphic chips should also include glial cells, given the proven importance of astrocytes, a type of glial cell which contributes to information processing in the brain. The Deep Artificial Neuron-Astrocyte Networks (DANAN) could overcome the difficulties in architecture design, learning process and scalability of the current ML methods.

Molecular interaction fingerprint approaches for GPCR drug discovery

Protein-ligand interaction fingerprints (IFPs) are binary 1D representations of the 3D structure of protein-ligand complexes encoding the presence or absence of specific interactions between the binding pocket amino acids and the ligand. Various implementations of IFPs have been developed and successfully applied for post-processing molecular docking results for G Protein-Coupled Receptor (GPCR) ligand binding mode prediction and virtual ligand screening. Novel interaction fingerprint methods enable structural chemogenomics and polypharmacology predictions by complementing the increasing amount of GPCR structural data. Machine learning methods are increasingly used to derive relationships between bioactivity data and fingerprint descriptors of chemical and structural information of binding sites, ligands, and protein-ligand interactions. Factors that influence the application of IFPs include structure preparation, binding site definition, fingerprint similarity assessment, and data processing and these factors pose challenges as well possibilities to optimize interaction fingerprint methods for GPCR drug discovery.