Bayer's in silico ADMET platform: a journey of machine learning over the past two decades.

Over the past two decades, an in silico absorption, distribution, metabolism, and excretion (ADMET) platform has been created at Bayer Pharma with the goal to generate models for a variety of pharmacokinetic and physicochemical endpoints in early drug discovery. These tools are accessible to all scientists within the company and can be a useful in assisting with the selection and design of novel leads, as well as the process of lead optimization. Here. we discuss the development of machine-learning (ML) approaches with special emphasis on data, descriptors, and algorithms. We show that high company internal data quality and tailored descriptors, as well as a thorough understanding of the experimental endpoints, are essential to the utility of our models. We discuss the recent impact of deep neural networks and show selected application examples.

Introducing the concept of virtual control groups into preclinical toxicology testing.

Sharing legacy data from in vivo toxicity studies offers the opportunity to analyze the variability of control groups stratified for strain, age, duration of study, vehicle and other experimental conditions. Historical animal control group data may lead to a repository, which could be used to construct virtual control groups (VCGs) for toxicity studies. VCGs are an established concept in clinical trials, but the idea of replacing living beings with virtual data sets has so far not been introduced into the design of regulatory animal studies. The use of VCGs has the potential of a 25% reduction in animal use by replacing the control group animals with existing randomized data sets. Prerequisites for such an approach are the availability of large and well-structured control data sets as well as thorough statistical evaluations. the foundation of data sharing has been laid within the Innovative Medicines Initiatives projects eTOX and eTRANSAFE. For a proof of principle participating companies have started to collect control group data for subacute (4-week) GLP studies with Wistar rats (the strain preferentially used in Europe) and are characterizing these data for its variability. In a second step, the control group data will be shared among the companies and cross-company variability will be investigated. In a third step, a set of studies will be analyzed to assess whether the use of VCG data would have influenced the outcome of the study compared to the real control group.

Genotoxicity Assessment of Drug Metabolites in the Context of MIST and Beyond.

While there are dedicated guidelines for industry regarding the assessment of the genotoxic potential of new pharmaceuticals and impurities, and the general safety assessment of major drug metabolites, only limited guidance exists on the assessment of potential genotoxic minor drug metabolites. In this Perspective, we discuss challenges associated with assessing the genotoxic potential of human metabolites and share five case studies within the context of an "aware-avoid-assess" paradigm. A special focus is on a class of potentially genotoxic carcinogens, aromatic amines (arylamines and anilines). This compound class is frequently used as building blocks and may show up as impurities, metabolites, or degradants in pharmaceuticals. We propose several recommendations that should help project teams at different stages of pharmaceutical development. In most cases, proactive interactions with the relevant health authority should be considered to endorse the proposed genotoxicity assessment strategy for minor drug metabolites.

Comprehensive and Automated Linear Interaction Energy Based Binding-Affinity Prediction for Multifarious Cytochrome P450 Aromatase Inhibitors.

Cytochrome P450 aromatase (CYP19A1) plays a key role in the development of estrogen dependent breast cancer, and aromatase inhibitors have been at the front line of treatment for the past three decades. The development of potent, selective and safer inhibitors is ongoing with in silico screening methods playing a more prominent role in the search for promising lead compounds in bioactivity-relevant chemical space. Here we present a set of comprehensive binding affinity prediction models for CYP19A1 using our automated Linear Interaction Energy (LIE) based workflow on a set of 132 putative and structurally diverse aromatase inhibitors obtained from a typical industrial screening study. We extended the workflow with machine learning methods to automatically cluster training and test compounds in order to maximize the number of explained compounds in one or more predictive LIE models. The method uses protein-ligand interaction profiles obtained from Molecular Dynamics (MD) trajectories to help model search and define the applicability domain of the resolved models. Our method was successful in accounting for 86% of the data set in 3 robust models that show high correlation between calculated and observed values for ligand-binding free energies (RMSE < 2.5 kJ mol-1), with good cross-validation statistics.

In silico prediction of genotoxicity.

The in silico prediction of genotoxicity has made considerable progress during the last years. The main driver for the pharmaceutical industry is the ICH M7 guideline about the assessment of DNA reactive impurities. An important component of this guideline is the use of in silico models as an alternative approach to experimental testing. The in silico prediction of genotoxicity provides an established and accepted method that defines the first step in the assessment of DNA reactive impurities. This was made possible by the growing amount of reliable Ames screening data, the attempts to understand the activity pathways and the subsequent development of computer-based prediction systems. This paper gives an overview of how the in silico prediction of genotoxicity is performed under the ICH M7 guideline.

The eTOX data-sharing project to advance in silico drug-induced toxicity prediction.

The high-quality in vivo preclinical safety data produced by the pharmaceutical industry during drug development, which follows numerous strict guidelines, are mostly not available in the public domain. These safety data are sometimes published as a condensed summary for the few compounds that reach the market, but the majority of studies are never made public and are often difficult to access in an automated way, even sometimes within the owning company itself. It is evident from many academic and industrial examples, that useful data mining and model development requires large and representative data sets and careful curation of the collected data. In 2010, under the auspices of the Innovative Medicines Initiative, the eTOX project started with the objective of extracting and sharing preclinical study data from paper or pdf archives of toxicology departments of the 13 participating pharmaceutical companies and using such data for establishing a detailed, well-curated database, which could then serve as source for read-across approaches (early assessment of the potential toxicity of a drug candidate by comparison of similar structure and/or effects) and training of predictive models. The paper describes the efforts undertaken to allow effective data sharing intellectual property (IP) protection and set up of adequate controlled vocabularies) and to establish the database (currently with over 4000 studies contributed by the pharma companies corresponding to more than 1400 compounds). In addition, the status of predictive models building and some specific features of the eTOX predictive system (eTOXsys) are presented as decision support knowledge-based tools for drug development process at an early stage.

Molecular evolution of a peptide GPCR ligand driven by artificial neural networks.

Peptide ligands of G protein-coupled receptors constitute valuable natural lead structures for the development of highly selective drugs and high-affinity tools to probe ligand-receptor interaction. Currently, pharmacological and metabolic modification of natural peptides involves either an iterative trial-and-error process based on structure-activity relationships or screening of peptide libraries that contain many structural variants of the native molecule. Here, we present a novel neural network architecture for the improvement of metabolic stability without loss of bioactivity. In this approach the peptide sequence determines the topology of the neural network and each cell corresponds one-to-one to a single amino acid of the peptide chain. Using a training set, the learning algorithm calculated weights for each cell. The resulting network calculated the fitness function in a genetic algorithm to explore the virtual space of all possible peptides. The network training was based on gradient descent techniques which rely on the efficient calculation of the gradient by back-propagation. After three consecutive cycles of sequence design by the neural network, peptide synthesis and bioassay this new approach yielded a ligand with 70fold higher metabolic stability compared to the wild type peptide without loss of the subnanomolar activity in the biological assay. Combining specialized neural networks with an exploration of the combinatorial amino acid sequence space by genetic algorithms represents a novel rational strategy for peptide design and optimization.

Inroads to predict in vivo toxicology-an introduction to the eTOX Project.

There is a widespread awareness that the wealth of preclinical toxicity data that the pharmaceutical industry has generated in recent decades is not exploited as efficiently as it could be. Enhanced data availability for compound comparison ("read-across"), or for data mining to build predictive tools, should lead to a more efficient drug development process and contribute to the reduction of animal use (3Rs principle). In order to achieve these goals, a consortium approach, grouping numbers of relevant partners, is required. The eTOX ("electronic toxicity") consortium represents such a project and is a public-private partnership within the framework of the European Innovative Medicines Initiative (IMI). The project aims at the development of in silico prediction systems for organ and in vivo toxicity. The backbone of the project will be a database consisting of preclinical toxicity data for drug compounds or candidates extracted from previously unpublished, legacy reports from thirteen European and European operation-based pharmaceutical companies. The database will be enhanced by incorporation of publically available, high quality toxicology data. Seven academic institutes and five small-to-medium size enterprises (SMEs) contribute with their expertise in data gathering, database curation, data mining, chemoinformatics and predictive systems development. The outcome of the project will be a predictive system contributing to early potential hazard identification and risk assessment during the drug development process. The concept and strategy of the eTOX project is described here, together with current achievements and future deliverables.

Chemogenomic analysis of G-protein coupled receptors and their ligands deciphers locks and keys governing diverse aspects of signalling.

Understanding the molecular mechanism of signalling in the important super-family of G-protein-coupled receptors (GPCRs) is causally related to questions of how and where these receptors can be activated or inhibited. In this context, it is of great interest to unravel the common molecular features of GPCRs as well as those related to an active or inactive state or to subtype specific G-protein coupling. In our underlying chemogenomics study, we analyse for the first time the statistical link between the properties of G-protein-coupled receptors and GPCR ligands. The technique of mutual information (MI) is able to reveal statistical inter-dependence between variations in amino acid residues on the one hand and variations in ligand molecular descriptors on the other. Although this MI analysis uses novel information that differs from the results of known site-directed mutagenesis studies or published GPCR crystal structures, the method is capable of identifying the well-known common ligand binding region of GPCRs between the upper part of the seven transmembrane helices and the second extracellular loop. The analysis shows amino acid positions that are sensitive to either stimulating (agonistic) or inhibitory (antagonistic) ligand effects or both. It appears that amino acid positions for antagonistic and agonistic effects are both concentrated around the extracellular region, but selective agonistic effects are cumulated between transmembrane helices (TMHs) 2, 3, and ECL2, while selective residues for antagonistic effects are located at the top of helices 5 and 6. Above all, the MI analysis provides detailed indications about amino acids located in the transmembrane region of these receptors that determine G-protein signalling pathway preferences.

Design of chemical libraries with potentially bioactive molecules applying a maximum common substructure concept.

Success in small molecule screening relies heavily on the preselection of compounds. Here, we present a strategy for the enrichment of chemical libraries with potentially bioactive compounds integrating the collected knowledge of medicinal chemistry. Employing a genetic algorithm, substructures typically occurring in bioactive compounds were identified using the World Drug Index. Availability of compounds containing the selected substructures was analysed in vendor libraries, and the substructure-specific sublibraries were assembled. Compounds containing reactive, undesired functional groups were omitted. Using a diversity filter for both physico-chemical properties and the substructure composition, the compounds of all the sublibraries were ranked. Accordingly, a screening collection of 16,671 compounds was selected. Diversity and chemical space coverage of the collection indicate that it is highly diverse and well-placed in the chemical space spanned by bioactive compounds. Furthermore, secondary assay-validated hits presented in this study show the practical relevance of our library design strategy.

Computational prediction of the chromosome-damaging potential of chemicals.

We report on the generation of computer-based models for the prediction of the chromosome-damaging potential of chemicals as assessed in the in vitro chromosome aberration (CA) test. On the basis of publicly available CA-test results of more than 650 chemical substances, half of which are drug-like compounds, we generated two different computational models. The first model was realized using the (Q)SAR tool MCASE. Results obtained with this model indicate a limited performance (53%) for the assessment of a chromosome-damaging potential (sensitivity), whereas CA-test negative compounds were correctly predicted with a specificity of 75%. The low sensitivity of this model might be explained by the fact that the underlying 2D-structural descriptors only describe part of the molecular mechanism leading to the induction of chromosome aberrations, that is, direct drug-DNA interactions. The second model was constructed with a more sophisticated machine learning approach and generated a classification model based on 14 molecular descriptors, which were obtained after feature selection. The performance of this model was superior to the MCASE model, primarily because of an improved sensitivity, suggesting that the more complex molecular descriptors in combination with statistical learning approaches are better suited to model the complex nature of mechanisms leading to a positive effect in the CA-test. An analysis of misclassified pharmaceuticals by this model showed that a large part of the false-negative predicted compounds were uniquely positive in the CA-test but lacked a genotoxic potential in other mutagenicity tests of the regulatory testing battery, suggesting that biologically nonsignificant mechanisms could be responsible for the observed positive CA-test result. Since such mechanisms are not amenable to modeling approaches it is suggested that a positive prediction made by the model reflects a biologically significant genotoxic potential. An integration of the machine-learning model as a screening tool in early discovery phases of drug development is proposed.