Matched Molecular Pair Analysis in Short: Algorithms, Applications and Limitations

OptADMET: a web-based tool for substructure modifications to improve ADMET properties of lead compounds

Lead optimization is a crucial step in the drug discovery process, which aims to design potential drug candidates from biologically active hits. During lead optimization, active hits undergo modifications to improve their absorption, distribution, metabolism, excretion and toxicity (ADMET) profiles. Medicinal chemists face key questions regarding which compound(s) should be synthesized next and how to balance multiple ADMET properties. Reliable transformation rules from multiple experimental analyses are critical to improve this decision-making process. We developed OptADMET ( https://cadd.nscc-tj.cn/deploy/optadmet/ ), an integrated web-based platform that provides chemical transformation rules for 32 ADMET properties and leverages prior experimental data for lead optimization. The multiproperty transformation rule database contains a total of 41,779 validated transformation rules generated from the analysis of 177,191 reliable experimental datasets. Additionally, 146,450 rules were generated by analyzing 239,194 molecular data predictions. OptADMET provides the ADMET profiles of all optimized molecules from the queried molecule and enables the prediction of desirable substructure transformations and subsequent validation of drug candidates. OptADMET is based on matched molecular pairs analysis derived from synthetic chemistry, thus providing improved practicality over other methods. OptADMET is designed for use by both experimental and computational scientists.

Predicting polypharmacy in half a million adults in the Iranian population: comparison of machine learning algorithms

BACKGROUND

Polypharmacy (PP) is increasingly common in Iran, and contributes to the substantial burden of drug-related morbidity, increasing the potential for drug interactions and potentially inappropriate medications. Machine learning algorithms (ML) can be employed as an alternative solution for the prediction of PP. Therefore, our study aimed to compare several ML algorithms to predict the PP using the health insurance claims data and choose the best-performing algorithm as a predictive tool for decision-making.

METHODS

This population-based cross-sectional study was performed between April 2021 and March 2022. After feature selection, information about 550 thousand patients were obtained from National Center for Health Insurance Research (NCHIR). Afterwards, several ML algorithms were trained to predict PP. Finally, to assess the models' performance, the metrics derived from the confusion matrix were calculated.

RESULTS

The study sample comprised 554 133 adults with a median (IQR) age of 51 years (40 - 62) that nested in 27 cities within the Khuzestan province of Iran. Most of the patients were female (62.5%), married (63.5%), and employed (83.2%) during the last year. The prevalence of PP in all populations was about 36.0%. After performing the feature selection, out of 23 features, the number of prescriptions, Insurance coverage for prescription drugs, and hypertension were found as the top three predictors. Experimental results showed that Random Forest (RF) performed better than other ML algorithms with recall, specificity, accuracy, precision and F1-score of 63.92%, 89.92%, 79.99%, 63.92% and 63.92% respectively.

CONCLUSION

It was found that ML provides a reasonable level of accuracy in predicting polypharmacy. Therefore, the prediction models based on ML, especially the RF algorithm, performed better than other methods for predicting PP in Iranian people in terms of the performance criteria.

TargIDe: a machine-learning workflow for target identification of molecules with antibiofilm activity against Pseudomonas aeruginosa

Bacterial biofilms are a source of infectious human diseases and are heavily linked to antibiotic resistance. Pseudomonas aeruginosa is a multidrug-resistant bacterium widely present and implicated in several hospital-acquired infections. Over the last years, the development of new drugs able to inhibit Pseudomonas aeruginosa by interfering with its ability to form biofilms has become a promising strategy in drug discovery. Identifying molecules able to interfere with biofilm formation is difficult, but further developing these molecules by rationally improving their activity is particularly challenging, as it requires knowledge of the specific protein target that is inhibited. This work describes the development of a machine learning multitechnique consensus workflow to predict the protein targets of molecules with confirmed inhibitory activity against biofilm formation by Pseudomonas aeruginosa. It uses a specialized database containing all the known targets implicated in biofilm formation by Pseudomonas aeruginosa. The experimentally confirmed inhibitors available on ChEMBL, together with chemical descriptors, were used as the input features for a combination of nine different classification models, yielding a consensus method to predict the most likely target of a ligand. The implemented algorithm is freely available at https://github.com/BioSIM-Research-Group/TargIDe under licence GNU General Public Licence (GPL) version 3 and can easily be improved as more data become available.

Matcher: An Open-Source Application for Translating Large Structure/Property Data Sets into Insights for Drug Design

To solve recurring problems in drug discovery, matched molecular pair (MMP) analysis is used to understand relationships between chemical structure and function. For the MMP analysis of large data sets (>10,000 compounds), available tools lack flexible search and visualization functionality and require computational expertise. Here, we present Matcher, an open-source application for MMP analysis, with novel search algorithms and fully automated querying-to-visualization that requires no programming expertise. Matcher enables unprecedented control over the search and clustering of MMP transformations based on both variable fragment and constant environment structure, which is critical for disentangling relevant and irrelevant data to a given problem. Users can exert such control through a built-in chemical sketcher and with a few mouse clicks can navigate between resulting MMP transformations, statistics, property distribution graphs, and structures with raw experimental data, for confident and accelerated decision making. Matcher can be used with any collection of structure/property data; here, we demonstrate usage with a public ChEMBL data set of about 20,000 small molecules with CYP3A4 and/or hERG inhibition data. Users can reproduce all examples demonstrated herein via unique links within Matcher's interface-a functionality that anyone can use to preserve and share their own analyses. Matcher and all its dependencies are open-source, can be used for free, and are available with containerized deployment from code at https://github.com/Merck/Matcher. Matcher makes large structure/property data sets more transparent than ever before and accelerates the data-driven solution of common problems in drug discovery.

A Perspective on Explanations of Molecular Prediction Models

Chemists can be skeptical in using deep learning (DL) in decision making, due to the lack of interpretability in "black-box" models. Explainable artificial intelligence (XAI) is a branch of artificial intelligence (AI) which addresses this drawback by providing tools to interpret DL models and their predictions. We review the principles of XAI in the domain of chemistry and emerging methods for creating and evaluating explanations. Then, we focus on methods developed by our group and their applications in predicting solubility, blood-brain barrier permeability, and the scent of molecules. We show that XAI methods like chemical counterfactuals and descriptor explanations can explain DL predictions while giving insight into structure-property relationships. Finally, we discuss how a two-step process of developing a black-box model and explaining predictions can uncover structure-property relationships.

Can We Quickly Learn to "Translate" Bioactive Molecules with Transformer Models?

Meaningful exploration of the chemical space of druglike molecules in drug design is a highly challenging task due to a combinatorial explosion of possible modifications of molecules. In this work, we address this problem with transformer models, a type of machine learning (ML) model originally developed for machine translation. By training transformer models on pairs of similar bioactive molecules from the public ChEMBL data set, we enable them to learn medicinal-chemistry-meaningful, context-dependent transformations of molecules, including those absent from the training set. By retrospective analysis on the performance of transformer models on ChEMBL subsets of ligands binding to COX2, DRD2, or HERG protein targets, we demonstrate that the models can generate structures identical or highly similar to most active ligands, despite the models having not seen any ligands active against the corresponding protein target during training. Our work demonstrates that human experts working on hit expansion in drug design can easily and quickly employ transformer models, originally developed to translate texts from one natural language to another, to "translate" from known molecules active against a given protein target to novel molecules active against the same target.

Computer-aided multi-objective optimization in small molecule discovery

Molecular discovery is a multi-objective optimization problem that requires identifying a molecule or set of molecules that balance multiple, often competing, properties. Multi-objective molecular design is commonly addressed by combining properties of interest into a single objective function using scalarization, which imposes assumptions about relative importance and uncovers little about the trade-offs between objectives. In contrast to scalarization, Pareto optimization does not require knowledge of relative importance and reveals the trade-offs between objectives. However, it introduces additional considerations in algorithm design. In this review, we describe pool-based and de novo generative approaches to multi-objective molecular discovery with a focus on Pareto optimization algorithms. We show how pool-based molecular discovery is a relatively direct extension of multi-objective Bayesian optimization and how the plethora of different generative models extend from single-objective to multi-objective optimization in similar ways using non-dominated sorting in the reward function (reinforcement learning) or to select molecules for retraining (distribution learning) or propagation (genetic algorithms). Finally, we discuss some remaining challenges and opportunities in the field, emphasizing the opportunity to adopt Bayesian optimization techniques into multi-objective de novo design.

Transformer-based deep learning method for optimizing ADMET properties of lead compounds

A successful drug needs to exhibit both effective pharmacodynamics (PD) and safe pharmacokinetics (PK). However, the coordinated optimization of PD and PK properties in molecule generation tasks remains a great challenge for most existing methods, especially when they focus on the pursuit of affinity and selectivity for the lead compound. Thus, molecular optimization for PK properties is a critical step in the drug discovery pipeline, in which absorption, distribution, metabolism, excretion and toxicity (ADMET) property predictive models play an increasingly important role by providing an effective method to assess multiple PK properties of compounds. Here, we proposed a Graph Bert-based ADMET prediction model that achieves state-of-the-art performance on the public dataset Therapeutics Data Commons (TDC) by combining molecular graph features and descriptor features, with 11 tasks ranked first and 20 tasks ranked in the top 3. Based on this prediction model, we trained a Transformer model with multiple properties as constraints for learning the structural transformations involved in MMP and the accompanying property changes. The experimental results show that the trained Constraints-Transformer can implement targeted modifications to the starting molecule, while preserving the core scaffold. Moreover, molecular docking and binding mode analysis demonstrate that the optimized molecules still retain the activity and selectivity for biological targets. Therefore, the proposed method accounts for biological activity and ADMET properties simultaneously. Finally, a webserver containing ADMET property prediction and molecular optimization functions is provided, enabling chemists to improve the properties of starting molecules individually.

BIreactive: Expanding the Scope of Reactivity Predictions to Propynamides

We present the first comprehensive study on the prediction of reactivity for propynamides. Covalent inhibitors like propynamides often show improved potency, selectivity, and unique pharmacologic properties compared to their non-covalent counterparts. In order to achieve this, it is essential to tune the reactivity of the warhead. This study shows how three different in silico methods can predict the in vitro properties of propynamides, a covalent warhead class integrated into approved drugs on the market. Whereas the electrophilicity index is only applicable to individual subclasses of substitutions, adduct formation and transition state energies have a good predictability for the in vitro reactivity with glutathione (GSH). In summary, the reported methods are well suited to estimate the reactivity of propynamides. With this knowledge, the fine tuning of the reactivity is possible which leads to a speed up of the design process of covalent drugs.

KID: A Kinase-Focused Interaction Database and Its Application in the Construction of Kinase-Focused Molecule Databases

Protein kinases are important drug targets for the treatment of several diseases. The interaction between kinases and ligands is vital in the process of small-molecule kinase inhibitor (SMKI) design. In this study, we propose a method to extract fragments and amino acid residues from crystal structures for kinase-ligand interactions. In addition, core fragments that interact with the important hinge region of kinases were extracted along with their decorations. Based on the superimposed structural data of kinases from the kinase-ligand interaction fingerprint and structure database, we obtained two libraries, namely, a hinge-unfocused fragment-amino acid pair library (FAP Lib) that contains 6672 pairs of fragments and corresponding amino-acids, and a hinge-focused hinge binder library (HB Lib) of 3560 pairs of hinge-binding scaffolds with their corresponding decorations. These two libraries constitute a kinase-focused interaction database (KID). In depth analysis was conducted on KID to explore important characteristics of fragments in the design of SMKIs. With KID, we built two kinase-focused molecule databases, one called Recomb_DB, which contains 1,72,346 molecules generated through fragment recombination based on the FAP Lib, and another called RsdHB_DB, which contains 93,030 molecules generated based on our HB Lib using molecular generation methods. Compared with five databases both commercial and non-commercial, these two databases both ranked top 3 in scaffold diversity, top 4 in molecule fingerprint diversity, and are more focused on the chemical space of kinase inhibitors. Hence, KID presents a useful addition to existing databases for the exploration of novel SMKIs.

Natural and Synthetic Xanthone Derivatives Counteract Oxidative Stress via Nrf2 Modulation in Inflamed Human Macrophages

Natural products have attracted attention due to their safety and potential effectiveness as anti-inflammatory drugs. Particularly, xanthones owning a unique 9H-xanthen-9-one scaffold, are endowed with a large diversity of medical applications, including antioxidant and anti-inflammatory activities, because their core accommodates a vast variety of substituents at different positions. Among others, α- and γ-mangostin are the major known xanthones purified from Garcinia mangostana with demonstrated anti-inflammatory and antioxidant effects by in vitro and in vivo modulation of the Nrf2 (nuclear factor erythroid-derived 2-like 2) pathway. However, the main mechanism of action of xanthones and their derivatives is still only partially disclosed, and further investigations are needed to improve their potential clinical outcomes. In this light, a library of xanthone derivatives was synthesized and biologically evaluated in vitro on human macrophages under pro-inflammatory conditions. Furthermore, structure-activity relationship (SAR) studies were performed by means of matched molecular pairs (MMPs). The data obtained revealed that the most promising compounds in terms of biocompatibility and counteraction of cytotoxicity are the ones that enhance the Nrf2 translocation, confirming a tight relationship between the xanthone scaffold and the Nrf2 activation as a sign of intracellular cell response towards oxidative stress and inflammation.

Implications of Additivity and Nonadditivity for Machine Learning and Deep Learning Models in Drug Design

Matched molecular pairs (MMPs) are nowadays a commonly applied concept in drug design. They are used in many computational tools for structure-activity relationship analysis, biological activity prediction, or optimization of physicochemical properties. However, until now it has not been shown in a rigorous way that MMPs, that is, changing only one substituent between two molecules, can be predicted with higher accuracy and precision in contrast to any other chemical compound pair. It is expected that any model should be able to predict such a defined change with high accuracy and reasonable precision. In this study, we examine the predictability of four classical properties relevant for drug design ranging from simple physicochemical parameters (log D and solubility) to more complex cell-based ones (permeability and clearance), using different data sets and machine learning algorithms. Our study confirms that additive data are the easiest to predict, which highlights the importance of recognition of nonadditivity events and the challenging complexity of predicting properties in case of scaffold hopping. Despite deep learning being well suited to model nonlinear events, these methods do not seem to be an exception of this observation. Though they are in general performing better than classical machine learning methods, this leaves the field with a still standing challenge.

Machine Learning Applications for Chemical Reactions

Machine learning (ML) approaches have enabled rapid and efficient molecular property predictions as well as the design of new novel materials. In addition to great success for molecular problems, ML techniques are applied to various chemical reaction problems that require huge costs to solve with the existing experimental and simulation methods. In this review, starting with basic representations of chemical reactions, we summarized recent achievements of ML studies on two different problems; predicting reaction properties and synthetic routes. The various ML models are used to predict physical properties related to chemical reaction properties (e. g. thermodynamic changes, activation barriers, and reaction rates). Furthermore, the predictions of reactivity, self-optimization of reaction, and designing retrosynthetic reaction paths are also tackled by ML approaches. Herein we illustrate various ML strategies utilized in the various context of chemical reaction studies.

IDL-PPBopt: A Strategy for Prediction and Optimization of Human Plasma Protein Binding of Compounds via an Interpretable Deep Learning Method

The prediction and optimization of pharmacokinetic properties are essential in lead optimization. Traditional strategies mainly depend on the empirical chemical rules from medicinal chemists. However, with the rising amount of data, it is getting more difficult to manually extract useful medicinal chemistry knowledge. To this end, we introduced IDL-PPBopt, a computational strategy for predicting and optimizing the plasma protein binding (PPB) property based on an interpretable deep learning method. At first, a curated PPB data set was used to construct an interpretable deep learning model, which showed excellent predictive performance with a root mean squared error of 0.112 for the entire test set. Then, we designed a detection protocol based on the model and Wilcoxon test to identify the PPB-related substructures (named privileged substructures, PSubs) for each molecule. In total, 22 general privileged substructures (GPSubs) were identified, which shared some common features such as nitrogen-containing groups, diamines with two carbon units, and azetidine. Furthermore, a series of second-level chemical rules for each GPSub were derived through a statistical test and then summarized into substructure pairs. We demonstrated that these substructure pairs were equally applicable outside the training set and accordingly customized the structural modification schemes for each GPSub, which provided alternatives for the optimization of the PPB property. Therefore, IDL-PPBopt provides a promising scheme for the prediction and optimization of the PPB property and would be helpful for lead optimization of other pharmacokinetic properties.

Application of Artificial Intelligence in Discovery and Development of Anticancer and Antidiabetic Therapeutic Agents

Spectacular developments in molecular and cellular biology have led to important discoveries in cancer research. Despite cancer is one of the major causes of morbidity and mortality globally, diabetes is one of the most leading sources of group of disorders. Artificial intelligence (AI) has been considered the fourth industrial revolution machine. The most major hurdles in drug discovery and development are the time and expenditures required to sustain the drug research pipeline. Large amounts of data can be explored and generated by AI, which can then be converted into useful knowledge. Because of this, the world's largest drug companies have already begun to use AI in their drug development research. In the present era, AI has a huge amount of potential for the rapid discovery and development of new anticancer drugs. Clinical studies, electronic medical records, high-resolution medical imaging, and genomic assessments are just a few of the tools that could aid drug development. Large data sets are available to researchers in the pharmaceutical and medical fields, which can be analyzed by advanced AI systems. This review looked at how computational biology and AI technologies may be utilized in cancer precision drug development by combining knowledge of cancer medicines, drug resistance, and structural biology. This review also highlighted a realistic assessment of the potential for AI in understanding and managing diabetes.

Transformer-based molecular optimization beyond matched molecular pairs

Molecular optimization aims to improve the drug profile of a starting molecule. It is a fundamental problem in drug discovery but challenging due to (i) the requirement of simultaneous optimization of multiple properties and (ii) the large chemical space to explore. Recently, deep learning methods have been proposed to solve this task by mimicking the chemist’s intuition in terms of matched molecular pairs (MMPs). Although MMPs is a widely used strategy by medicinal chemists, it offers limited capability in terms of exploring the space of structural modifications, therefore does not cover the complete space of solutions. Often more general transformations beyond the nature of MMPs are feasible and/or necessary, e.g. simultaneous modifications of the starting molecule at different places including the core scaffold. This study aims to provide a general methodology that offers more general structural modifications beyond MMPs. In particular, the same Transformer architecture is trained on different datasets. These datasets consist of a set of molecular pairs which reflect different types of transformations. Beyond MMP transformation, datasets reflecting general structural changes are constructed from ChEMBL based on two approaches: Tanimoto similarity (allows for multiple modifications) and scaffold matching (allows for multiple modifications but keep the scaffold constant) respectively. We investigate how the model behavior can be altered by tailoring the dataset while using the same model architecture. Our results show that the models trained on differently prepared datasets transform a given starting molecule in a way that it reflects the nature of the dataset used for training the model. These models could complement each other and unlock the capability for the chemists to pursue different options for improving a starting molecule.

Benchmarking Molecular Feature Attribution Methods with Activity Cliffs

Feature attribution techniques are popular choices within the explainable artificial intelligence toolbox, as they can help elucidate which parts of the provided inputs used by an underlying supervised-learning method are considered relevant for a specific prediction. In the context of molecular design, these approaches typically involve the coloring of molecular graphs, whose presentation to medicinal chemists can be useful for making a decision of which compounds to synthesize or prioritize. The consistency of the highlighted moieties alongside expert background knowledge is expected to contribute to the understanding of machine-learning models in drug design. Quantitative evaluation of such coloring approaches, however, has so far been limited to substructure identification tasks. We here present an approach that is based on maximum common substructure algorithms applied to experimentally-determined activity cliffs. Using the proposed benchmark, we found that molecule coloring approaches in conjunction with classical machine-learning models tend to outperform more modern, graph-neural-network alternatives. The provided benchmark data are fully open sourced, which we hope will facilitate the testing of newly developed molecular feature attribution techniques.

Molecule Ideation Using Matched Molecular Pairs

Matched Molecular Pair Analysis (MMP) is a very important tool during the lead optimization stage in drug discovery. The usefulness of this tool in the lead optimization stage has been discussed in several peer-reviewed articles. The application of MMP in Molecule generation is relatively new. This brings several challenges one of them being the need to encode contextual information into the transforms. In this chapter, we discuss how we use MMPs as a molecule generation method and how does it compare with other molecular generators.

An Open Drug Discovery Competition: Experimental Validation of Predictive Models in a Series of Novel Antimalarials

The Open Source Malaria (OSM) consortium is developing compounds that kill the human malaria parasite, Plasmodium falciparum, by targeting PfATP4, an essential ion pump on the parasite surface. The structure of PfATP4 has not been determined. Here, we describe a public competition created to develop a predictive model for the identification of PfATP4 inhibitors, thereby reducing project costs associated with the synthesis of inactive compounds. Competition participants could see all entries as they were submitted. In the final round, featuring private sector entrants specializing in machine learning methods, the best-performing models were used to predict novel inhibitors, of which several were synthesized and evaluated against the parasite. Half possessed biological activity, with one featuring a motif that the human chemists familiar with this series would have dismissed as "ill-advised". Since all data and participant interactions remain in the public domain, this research project "lives" and may be improved by others.

Recent omics-based computational methods for COVID-19 drug discovery and repurposing

The coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is the main reason for the increasing number of deaths worldwide. Although strict quarantine measures were followed in many countries, the disease situation is still intractable. Thus, it is needed to utilize all possible means to confront this pandemic. Therefore, researchers are in a race against the time to produce potential treatments to cure or reduce the increasing infections of COVID-19. Computational methods are widely proving rapid successes in biological related problems, including diagnosis and treatment of diseases. Many efforts in recent months utilized Artificial Intelligence (AI) techniques in the context of fighting the spread of COVID-19. Providing periodic reviews and discussions of recent efforts saves the time of researchers and helps to link their endeavors for a faster and efficient confrontation of the pandemic. In this review, we discuss the recent promising studies that used Omics-based data and utilized AI algorithms and other computational tools to achieve this goal. We review the established datasets and the developed methods that were basically directed to new or repurposed drugs, vaccinations and diagnosis. The tools and methods varied depending on the level of details in the available information such as structures, sequences or metabolic data.

Automatic Identification of Analogue Series from Large Compound Data Sets: Methods and Applications

Analogue series play a key role in drug discovery. They arise naturally in lead optimization efforts where analogues are explored based on one or a few core structures. However, it is much harder to accurately identify and extract pairs or series of analogue molecules in large compound databases with no predefined core structures. This methodological review outlines the most common and recent methodological developments to automatically identify analogue series in large libraries. Initial approaches focused on using predefined rules to extract scaffold structures, such as the popular Bemis-Murcko scaffold. Later on, the matched molecular pair concept led to efficient algorithms to identify similar compounds sharing a common core structure by exploring many putative scaffolds for each compound. Further developments of these ideas yielded, on the one hand, approaches for hierarchical scaffold decomposition and, on the other hand, algorithms for the extraction of analogue series based on single-site modifications (so-called matched molecular series) by exploring potential scaffold structures based on systematic molecule fragmentation. Eventually, further development of these approaches resulted in methods for extracting analogue series defined by a single core structure with several substitution sites that allow convenient representations, such as R-group tables. These methods enable the efficient analysis of large data sets with hundreds of thousands or even millions of compounds and have spawned many related methodological developments.

Interpretation of Ligand-Based Activity Cliff Prediction Models Using the Matched Molecular Pair Kernel

Activity cliffs (ACs) are formed by two structurally similar compounds with a large difference in potency. Accurate AC prediction is expected to help researchers' decisions in the early stages of drug discovery. Previously, predictive models based on matched molecular pair (MMP) cliffs have been proposed. However, the proposed methods face a challenge of interpretability due to the black-box character of the predictive models. In this study, we developed interpretable MMP fingerprints and modified a model-specific interpretation approach for models based on a support vector machine (SVM) and MMP kernel. We compared important features highlighted by this SVM-based interpretation approach and the SHapley Additive exPlanations (SHAP) as a major model-independent approach. The model-specific approach could capture the difference between AC and non-AC, while SHAP assigned high weights to the features not present in the test instances. For specific MMPs, the feature weights mapped by the SVM-based interpretation method were in agreement with the previously confirmed binding knowledge from X-ray co-crystal structures, indicating that this method is able to interpret the AC prediction model in a chemically intuitive manner.

Pairwise Difference Regression: A Machine Learning Meta-algorithm for Improved Prediction and Uncertainty Quantification in Chemical Search

Machine learning (ML) plays a growing role in the design and discovery of chemicals, aiming to reduce the need to perform expensive experiments and simulations. ML for such applications is promising but difficult, as models must generalize to vast chemical spaces from small training sets and must have reliable uncertainty quantification metrics to identify and prioritize unexplored regions. Ab initio computational chemistry and chemical intuition alike often take advantage of differences between chemical conditions, rather than their absolute structure or state, to generate more reliable results. We have developed an analogous comparison-based approach for ML regression, called pairwise difference regression (PADRE), which is applicable to arbitrary underlying learning models and operates on pairs of input data points. During training, the model learns to predict differences between all possible pairs of input points. During prediction, the test points are paired with all training set points, giving rise to a set of predictions that can be treated as a distribution of which the mean is treated as a final prediction and the dispersion is treated as an uncertainty measure. Pairwise difference regression was shown to reliably improve the performance of the random forest algorithm across five chemical ML tasks. Additionally, the pair-derived dispersion is both well correlated with model error and performs well in active learning. We also show that this method is competitive with state-of-the-art neural network techniques. Thus, pairwise difference regression is a promising tool for candidate selection algorithms used in chemical discovery.

Artificial intelligence to deep learning: machine intelligence approach for drug discovery

Drug designing and development is an important area of research for pharmaceutical companies and chemical scientists. However, low efficacy, off-target delivery, time consumption, and high cost impose a hurdle and challenges that impact drug design and discovery. Further, complex and big data from genomics, proteomics, microarray data, and clinical trials also impose an obstacle in the drug discovery pipeline. Artificial intelligence and machine learning technology play a crucial role in drug discovery and development. In other words, artificial neural networks and deep learning algorithms have modernized the area. Machine learning and deep learning algorithms have been implemented in several drug discovery processes such as peptide synthesis, structure-based virtual screening, ligand-based virtual screening, toxicity prediction, drug monitoring and release, pharmacophore modeling, quantitative structure-activity relationship, drug repositioning, polypharmacology, and physiochemical activity. Evidence from the past strengthens the implementation of artificial intelligence and deep learning in this field. Moreover, novel data mining, curation, and management techniques provided critical support to recently developed modeling algorithms. In summary, artificial intelligence and deep learning advancements provide an excellent opportunity for rational drug design and discovery process, which will eventually impact mankind. The primary concern associated with drug design and development is time consumption and production cost. Further, inefficiency, inaccurate target delivery, and inappropriate dosage are other hurdles that inhibit the process of drug delivery and development. With advancements in technology, computer-aided drug design integrating artificial intelligence algorithms can eliminate the challenges and hurdles of traditional drug design and development. Artificial intelligence is referred to as superset comprising machine learning, whereas machine learning comprises supervised learning, unsupervised learning, and reinforcement learning. Further, deep learning, a subset of machine learning, has been extensively implemented in drug design and development. The artificial neural network, deep neural network, support vector machines, classification and regression, generative adversarial networks, symbolic learning, and meta-learning are examples of the algorithms applied to the drug design and discovery process. Artificial intelligence has been applied to different areas of drug design and development process, such as from peptide synthesis to molecule design, virtual screening to molecular docking, quantitative structure-activity relationship to drug repositioning, protein misfolding to protein-protein interactions, and molecular pathway identification to polypharmacology. Artificial intelligence principles have been applied to the classification of active and inactive, monitoring drug release, pre-clinical and clinical development, primary and secondary drug screening, biomarker development, pharmaceutical manufacturing, bioactivity identification and physiochemical properties, prediction of toxicity, and identification of mode of action.

MolOpt: A Web Server for Drug Design using Bioisosteric Transformation

BACKGROUND

Bioisosteric replacement is widely used in drug design for lead optimization. However, the identification of a suitable bioisosteric group is not an easy task.

METHODS

In this work, we present MolOpt, a web server for in silico drug design using bioisosteric transformation. Potential bioisosteric transformation rules were derived from data mining, deep generative machine learning and similarity comparison. MolOpt tries to assist the medicinal chemist in his/her search for what to make next.

RESULTS AND DISCUSSION

By replacing molecular substructures with similar chemical groups, MolOpt automatically generates lists of analogues. MolOpt also evaluates forty important pharmacokinetic and toxic properties for each newly designed molecule. The transformed analogues can be assessed for possible future study.

CONCLUSION

MolOpt is useful for the identification of suitable lead optimization ideas. The MolOpt Server is freely available for use on the web at http://xundrug.cn/molopt.

Performance of Relative Binding Free Energy Calculations on an Automatically Generated Dataset of Halogen-Deshalogen Matched Molecular Pairs

In this study, we generated a matched molecular pair dataset of halogen/deshalogen compounds with reliable binding affinity data and structural binding mode information from public databases. The workflow includes automated system preparation and setup of free energy perturbation relative binding free energy calculations. We demonstrate the suitability of these datasets to investigate the performance of molecular mechanics force fields and molecular simulation algorithms for the purpose of in silico affinity predictions in lead optimization. Our datasets of a total of 115 matched molecular pairs show highly accurate binding free energy predictions with an average error of <1 kcal/mol despite the semi-automated calculation scheme. We quantify the accuracy of the optimized potential for liquid simulations (OPLS) force field to predict the effect of halogen addition to compounds, a commonly employed chemical modification in the design of drug-like molecules.

A matched molecular pair (MMP) approach for selecting analogs suitable for structure activity relationship (SAR)-based read across

One of the most challenging aspects of SAR-based read across is the identification of structurally similar compounds suitable for use as data sources to cover the safety of a target chemical. Matched molecular pair analysis (MMPA) provides a systematic method for mining experimental data for chemical substitutions that may be interpreted in terms of changes in properties. Here we use the relationships between structural substitutions linking a target chemical with an analog determined to be suitable using the expert-judgment based P&G framework of Wu et al. (2010). The relationships are established by applying MMPA to a database of compounds with safety assessed using SAR-based read across to suitable analogs possessing toxicological data. The analysis revealed that only five categories of substitutions per chemical class (aromatic or aliphatic) were necessary to link all molecular pairs. These data are summarized in a workflow outlining a strategy for searching toxicological databases for potential analogs. This approach provides structural comparisons that are interpretable and sensitive to small differences in the local structure of two compounds that may be linked to suitability for read across in contrast to the use of quantitative similarity measures which show little correlation with analog suitability.

Sub-structure-based screening and molecular docking studies of potential enteroviruses inhibitors

Rhinoviruses (RV), especially Human rhinovirus (HRVs) have been accepted as the most common cause for upper respiratory tract infections (URTIs). Pleconaril, a broad spectrum anti-rhinoviral compound, has been used as a drug of choice for URTIs for over a decade. Unfortunately, for various complications associated with this drug, it was rejected, and a replacement is highly desirable. In silico screening and prediction methods such as sub-structure search and molecular docking have been widely used to identify alternative compounds. In our study, we have utilised sub-structure search to narrow down our quest in finding relevant chemical compounds. Molecular docking studies were then used to study their binding interaction at the molecular level. Interestingly, we have identified 3 residues that is worth further investigation in upcoming molecular dynamics simulation systems of their contribution in stable interaction.

Molecular optimization by capturing chemist's intuition using deep neural networks

A main challenge in drug discovery is finding molecules with a desirable balance of multiple properties. Here, we focus on the task of molecular optimization, where the goal is to optimize a given starting molecule towards desirable properties. This task can be framed as a machine translation problem in natural language processing, where in our case, a molecule is translated into a molecule with optimized properties based on the SMILES representation. Typically, chemists would use their intuition to suggest chemical transformations for the starting molecule being optimized. A widely used strategy is the concept of matched molecular pairs where two molecules differ by a single transformation. We seek to capture the chemist’s intuition from matched molecular pairs using machine translation models. Specifically, the sequence-to-sequence model with attention mechanism, and the Transformer model are employed to generate molecules with desirable properties. As a proof of concept, three ADMET properties are optimized simultaneously: logD, solubility, and clearance, which are important properties of a drug. Since desirable properties often vary from project to project, the user-specified desirable property changes are incorporated into the input as an additional condition together with the starting molecules being optimized. Thus, the models can be guided to generate molecules satisfying the desirable properties. Additionally, we compare the two machine translation models based on the SMILES representation, with a graph-to-graph translation model HierG2G, which has shown the state-of-the-art performance in molecular optimization. Our results show that the Transformer can generate more molecules with desirable properties by making small modifications to the given starting molecules, which can be intuitive to chemists. A further enrichment of diverse molecules can be achieved by using an ensemble of models.

The Playbooks of Medicinal Chemistry Design Moves

Large databases of biologically relevant molecules, such as ChEMBL, SureChEMBL, or compound collections of pharmaceutical or agrochemical companies, are invaluable sources of medicinal chemistry information, albeit implicit. We developed a modified matched molecular pair approach to systematically and exhaustively extract the transformations in these databases and distill them into snippets of explicit design knowledge that are easily interpretable and directly applicable. The resulting "playbooks of medicinal chemistry design moves" capture the collective pharmaceutical and agrochemical research expertise across multiple chemists, companies, targets, and projects. They can be queried in an automated fashion for systematic prospective design and compound generation. The ChEMBL playbook and an application to exploit it are available at https://github.com/mahendra-awale/medchem_moves.

Design, synthesis and stepwise optimization of nitrile-based inhibitors of cathepsins B and L

Human cathepsin B (CatB) is an important biological target in cancer therapy. In this work, we performed a knowledge-based design approach and the synthesis of a new set of 19 peptide-like nitrile-based cathepsin inhibitors. Reported compounds were assayed against a panel of human cysteine proteases: CatB, CatL, CatK, and CatS. Three compounds (7h, 7i, and 7j) displayed nanomolar inhibition of CatB and selectivity over CatK and CatL. The selectivity was achieved by using the combination of a para biphenyl ring at P3, halogenated phenylalanine in P2 and Thr-O-Bz group at P1. Likewise, compounds 7i and 7j showed selective CatB inhibition among the panel of enzymes studied. We have also described a successful example of bioisosteric replacement of the amide bond for a sulfonamide one [7e → 6b], where we observed an increase in affinity and selectivity for CatB while lowering the compound lipophilicity (ilogP). Our knowledge-based design approach and the respective structure-activity relationships provide insights into the specific ligand-target interactions for therapeutically relevant cathepsins.

Retrospective assessment of rat liver microsomal stability at NCATS: data and QSAR models

Hepatic metabolic stability is a key pharmacokinetic parameter in drug discovery. Metabolic stability is usually assessed in microsomal fractions and only the best compounds progress in the drug discovery process. A high-throughput single time point substrate depletion assay in rat liver microsomes (RLM) is employed at the National Center for Advancing Translational Sciences. Between 2012 and 2020, RLM stability data was generated for ~ 24,000 compounds from more than 250 projects that cover a wide range of pharmacological targets and cellular pathways. Although a crucial endpoint, little or no data exists in the public domain. In this study, computational models were developed for predicting RLM stability using different machine learning methods. In addition, a retrospective time-split validation was performed, and local models were built for projects that performed poorly with global models. Further analysis revealed inherent medicinal chemistry knowledge potentially useful to chemists in the pursuit of synthesizing metabolically stable compounds. In addition, we deposited experimental data for ~ 2500 compounds in the PubChem bioassay database (AID: 1508591). The global prediction models are made publicly accessible ( https://opendata.ncats.nih.gov/adme ). This is to the best of our knowledge, the first publicly available RLM prediction model built using high-quality data generated at a single laboratory.

The Derivation of a Matched Molecular Pairs Based ADME/Tox Knowledge Base for Compound Optimization

Matched Molecular Pairs (MMP) analysis is a well-established technique for Structure Activity and Property Analysis (SAR and SPR). Summarizing multiple MMPs that describe the same structural change into a single chemical transform can be a powerful tool for prediction (termed Transform from here on). This is particularly useful in the area of Absorption, Distribution, Metabolism, and Elimination (ADME) analysis that is less influenced by 3D structural binding effects. The creation of a knowledge database containing many of these Transforms across typical ADME assays promises to be a powerful approach to aid multidimensional optimization. We present a detailed workflow for the derivation of such a database. We include details of an MMP fragmentation algorithm with associated statistical summarization methods for the derivation of Transforms. This is made freely available as part of the LillyMol software package. We describe the application of this method to several ADME/Tox (Toxicity) assay data sets and highlight multiple cases where the impact of traditional medicinal chemistry Transforms is contradicted by MMP data. We also describe the internal software interface used by medicinal chemists to aid the design of new compounds via automated suggestion. This approach utilizes the matched pairs database to "suggest" improved compounds in an automated design scenario. A nonvisual script-based version of the automated suggestions code with an associated set of described chemical Transforms is also made freely available along with this paper and as part of the LillyMol software package. Finally, we contrast this knowledge database against a larger database of all MMPs derived from a 2 million compound diversity set and a subset of MMPs seen in historical discovery projects. The comparison against all transforms in the diversity collection highlights the very low coverage of the transform database as compared to all possible transforms involving 15 atom fragments. The comparison against a smaller subset of Transforms seen on internal Medicinal Chemistry projects shows better coverage of the transform database for a small set of common medicinal chemistry strategies. Within the context of all possible transforms available to a medicinal chemistry project team, the challenge remains to move beyond mere idea generation from past projects toward high quality prediction for novel ADME/Tox modulating Transforms.

Tautomer Standardization in Chemical Databases: Deriving Business Rules from Quantum Chemistry

Databases of small, potentially bioactive molecules are ubiquitous across the industry and academia. Designed such that each unique compound should appear only once, the multiplicity of ways in which many compounds can be represented means that these databases require methods for standardizing the representation of chemistry. This is commonly achieved through the use of "Chemistry Business Rules", sets of predefined rules that describe the "house style" of the database in question. At Syngenta, the historical approach to the design of chemistry business rules has been to focus on consistency of representation, with chemical relevance given secondary consideration. In this work, we overturn that convention. Through the use of quantum chemistry calculations, we define a set of chemistry business rules for tautomer standardization that reproduces gas-phase energetic preferences. We go on to show that, compared to our historic approach, this method yields tautomers that are in better agreement with those observed experimentally in condensed phases and that are better suited for use in predictive models.

The Advent of Generative Chemistry

Generative adversarial networks (GANs), first published in 2014, are among the most important concepts in modern artificial intelligence (AI). Bridging deep learning and game theory, GANs are used to generate or "imagine" new objects with desired properties. Since 2016, multiple GANs with reinforcement learning (RL) have been successfully applied in pharmacology for de novo molecular design. Those techniques aim at a more efficient use of the data and a better exploration of the chemical space. We review recent advances for the generation of novel molecules with desired properties with a focus on the applications of GANs, RL, and related techniques. We also discuss the current limitations and challenges in the new growing field of generative chemistry.

Optimization strategy of single-digit nanomolar cross-class inhibitors of mammalian and protozoa cysteine proteases

Cysteine proteases (CPs) are involved in a myriad of actions that include not only protein degradation, but also play an essential biological role in infectious and systemic diseases such as cancer. CPs also act as biomarkers and can be reached by active-based probes for diagnostic and mechanistic purposes that are critical in health and disease. In this paper, we present the modulation of a CP panel of parasites and mammals (Trypanosoma cruzi cruzain, LmCPB, CatK, CatL and CatS), whose inhibition by nitrile peptidomimetics allowed the identification of specificity and selectivity for a given CP. The activity cliffs identified at the CP inhibition level are useful for retrieving trends through multiple structure-activity relationships. For two of the cruzain inhibitors (10g and 4e), both enthalpy and entropy are favourable to Gibbs binding energy, thus overcoming enthalpy-entropy compensation (EEC). Group contribution of individual molecular modification through changes in enthalpy and entropy results in a separate partition on the relative differences of Gibbs binding energy (ΔΔG). Overall, this study highlights the role of CPs in polypharmacology and multi-target screening, which represents an imperative trend in the actual drug discovery effort.

SMILES-based deep generative scaffold decorator for de-novo drug design

Molecular generative models trained with small sets of molecules represented as SMILES strings can generate large regions of the chemical space. Unfortunately, due to the sequential nature of SMILES strings, these models are not able to generate molecules given a scaffold (i.e., partially-built molecules with explicit attachment points). Herein we report a new SMILES-based molecular generative architecture that generates molecules from scaffolds and can be trained from any arbitrary molecular set. This approach is possible thanks to a new molecular set pre-processing algorithm that exhaustively slices all possible combinations of acyclic bonds of every molecule, combinatorically obtaining a large number of scaffolds with their respective decorations. Moreover, it serves as a data augmentation technique and can be readily coupled with randomized SMILES to obtain even better results with small sets. Two examples showcasing the potential of the architecture in medicinal and synthetic chemistry are described: First, models were trained with a training set obtained from a small set of Dopamine Receptor D2 (DRD2) active modulators and were able to meaningfully decorate a wide range of scaffolds and obtain molecular series predicted active on DRD2. Second, a larger set of drug-like molecules from ChEMBL was selectively sliced using synthetic chemistry constraints (RECAP rules). In this case, the resulting scaffolds with decorations were filtered only to allow those that included fragment-like decorations. This filtering process allowed models trained with this dataset to selectively decorate diverse scaffolds with fragments that were generally predicted to be synthesizable and attachable to the scaffold using known synthetic approaches. In both cases, the models were already able to decorate molecules using specific knowledge without the need to add it with other techniques, such as reinforcement learning. We envision that this architecture will become a useful addition to the already existent architectures for de novo molecular generation.

Matched Molecular Series Analysis for ADME Property Prediction

Generation and prioritization of new molecules are the most central part of the drug design process. Matched molecular series analysis (MMSA) has recently been proposed as a formal approach that captures both of these key elements of design. In order to better understand the power of MMSA and its specific limitations, we here evaluate its performance as an ADME property prediction tool. We use four large and diverse inhouse data sets, logD, microsomal clearance, CYP2C9, and CYP3A4 inhibition. MMSA follows the concept of parallel structure-activity relationship (SAR), where if two identical substituent series on different scaffolds show similarity in their property profiles, SAR from one series can be transferred to the other series. We test four different similarity metrics to identify pairs of molecular series where information can be transferred. We find that the best prediction performance is achieved by a combination of centered root-mean-square deviation (cRMSD) and a network score approach previously published by Keefer et al. However, cRMSD alone strikes the best balance between accuracy and the number of predictions that can be made. We identify statistical metrics that allow estimating when MMSA predictions will work, similar to the well-known applicability domain concept in machine learning. MMSA achieves a prediction accuracy that is comparable to a standard machine-learning model and matched molecular pair analysis. In contrast to machine learning, however, it is very easy to understand where MMSA predictions are coming from. Finally, to prospectively test the power of MMSA, we retested compounds that were strong outliers in the initial predictions and show how the MMSA model can help to identify erroneous data points.

Covalent fragment libraries in drug discovery

Targeted covalent inhibitors and chemical probes have become integral parts of drug discovery approaches. Given the advantages of fragment-based drug discovery, screening electrophilic fragments emerged as a promising alternative to discover and validate novel targets and to generate viable chemical starting points even for targets that are barely tractable. In this review, we present recent principles and considerations in the design of electrophilic fragment libraries from the selection of the appropriate covalent warhead through the design of the covalent fragment to the compilation of the library. We then summarize recent screening methodologies of covalent fragments against surrogate models, proteins, and the whole proteome, or living cells. Finally, we highlight recent drug discovery applications of covalent fragment libraries.

Mapping the S1 and S1' subsites of cysteine proteases with new dipeptidyl nitrile inhibitors as trypanocidal agents

The cysteine protease cruzipain is considered to be a validated target for therapeutic intervention in the treatment of Chagas disease. A series of 26 new compounds were designed, synthesized, and tested against the recombinant cruzain (Cz) to map its S1/S1´ subsites. The same series was evaluated on a panel of four human cysteine proteases (CatB, CatK, CatL, CatS) and Leishmania mexicana CPB, which is a potential target for the treatment of cutaneous leishmaniasis. The synthesized compounds are dipeptidyl nitriles designed based on the most promising combinations of different moieties in P1 (ten), P2 (six), and P3 (four different building blocks). Eight compounds exhibited a Ki smaller than 20.0 nM for Cz, whereas three compounds met these criteria for LmCPB. Three inhibitors had an EC50 value of ca. 4.0 μM, thus being equipotent to benznidazole according to the antitrypanosomal effects. Our mapping approach and the respective structure-activity relationships provide insights into the specific ligand-target interactions for therapeutically relevant cysteine proteases.

LogD Contributions of Substituents Commonly Used in Medicinal Chemistry

The importance of physicochemical property calculation and measurement is well-established in drug discovery. In particular, lipophilicity predictions play a central role in target design and prioritization. While significant progress has been made in our ability to calculate both logP and logD, the quality of these predictions is limited by the size and diversity of the underlying data set. Access to diverse data sets and advanced models is often limited to large organizations or consortia, and they are not available to many students and practitioners of medicinal chemistry. A molecular matched pair analysis of median ΔlogD _7.4 contributions for substituents commonly used in drug discovery programs at Genentech is reported. The results of this ΔlogD analysis are compiled into a single table, which we anticipate will be of use to practicing medicinal chemists.

Binding Kinetics Survey of the Drugged Kinome

Target residence time is emerging as an important optimization parameter in drug discovery, yet target and off-target engagement dynamics have not been clearly linked to the clinical performance of drugs. Here we developed high-throughput binding kinetics assays to characterize the interactions of 270 protein kinase inhibitors with 40 clinically relevant targets. Analysis of the results revealed that on-rates are better correlated with affinity than off-rates and that the fraction of slowly dissociating drug-target complexes increases from early/preclinical to late stage and FDA-approved compounds, suggesting distinct contributions by each parameter to clinical success. Combining binding parameters with PK/ADME properties, we illustrate in silico and in cells how kinetic selectivity could be exploited as an optimization strategy. Furthermore, using bio- and chemoinformatics we uncovered structural features influencing rate constants. Our results underscore the value of binding kinetics information in rational drug design and provide a resource for future studies on this subject.

Exploring Structure-Activity Relationships with Three-Dimensional Matched Molecular Pairs-A Review

A matched molecular pair (MMP) consists of two small molecules that differ by a few atoms only. The minor structural difference between the molecules allows a detailed analysis of changes in properties. Three-dimensional (3D) MMPs extend the concept of chemical similarity by spatial similarity. Conformations must be generated, and superimpositions have to be calculated. The additional complexity and uncertainty as well as the smaller amount of available experimental data substantially complicates the derivation of models. Nonetheless, there are some benefits that make the transition worthwhile. The 3D concept gives detailed insight into mechanisms behind several methods classically used by the 2D MMP approach. It can help to analyze disrupted series of structure-activity relationships or extend the 2D MMP concept with scaffold hopping. One of the most powerful features is the high confidence structure-activity relationship transfer between series of analogues. Several research groups have approached the problem from different directions. The models vary especially in the 3D similarity measure used and complexity of the applied descriptor selected or designed. Nonetheless, all approaches have increased the amount of information available by incorporating 3D structural information.