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.

Isoform-Selective ATAD2 Chemical Probe with Novel Chemical Structure and Unusual Mode of Action

ATAD2 (ANCCA) is an epigenetic regulator and transcriptional cofactor, whose overexpression has been linked to the progress of various cancer types. Here, we report a DNA-encoded library screen leading to the discovery of BAY-850, a potent and isoform selective inhibitor that specifically induces ATAD2 bromodomain dimerization and prevents interactions with acetylated histones in vitro, as well as with chromatin in cells. These features qualify BAY-850 as a chemical probe to explore ATAD2 biology.

Discovery and Characterization of a Highly Potent and Selective Aminopyrazoline-Based in Vivo Probe (BAY-598) for the Protein Lysine Methyltransferase SMYD2

Protein lysine methyltransferases have recently emerged as a new target class for the development of inhibitors that modulate gene transcription or signaling pathways. SET and MYND domain containing protein 2 (SMYD2) is a catalytic SET domain containing methyltransferase reported to monomethylate lysine residues on histone and nonhistone proteins. Although several studies have uncovered an important role of SMYD2 in promoting cancer by protein methylation, the biology of SMYD2 is far from being fully understood. Utilization of highly potent and selective chemical probes for target validation has emerged as a concept which circumvents possible limitations of knockdown experiments and, in particular, could result in an improved exploration of drug targets with a complex underlying biology. Here, we report the development of a potent, selective, and cell-active, substrate-competitive inhibitor of SMYD2, which is the first reported inhibitor suitable for in vivo target validation studies in rodents.

Accurate Solubility Prediction with Error Bars for Electrolytes: A Machine Learning Approach

Accurate in silico models for predicting aqueous solubility are needed in drug design and discovery and many other areas of chemical research. We present a statistical modeling of aqueous solubility based on measured data, using a Gaussian Process nonlinear regression model (GPsol). We compare our results with those of 14 scientific studies and 6 commercial tools. This shows that the developed model achieves much higher accuracy than available commercial tools for the prediction of solubility of electrolytes. On top of the high accuracy, the proposed machine learning model also provides error bars for each individual prediction.

Benzoisoquinolinediones as Potent and Selective Inhibitors of BRPF2 and TAF1/TAF1L Bromodomains

Bromodomains (BD) are readers of lysine acetylation marks present in numerous proteins associated with chromatin. Here we describe a dual inhibitor of the bromodomain and PHD finger (BRPF) family member BRPF2 and the TATA box binding protein-associated factors TAF1 and TAF1L. These proteins are found in large chromatin complexes and play important roles in transcription regulation. The substituted benzoisoquinolinedione series was identified by high-throughput screening, and subsequent structure–activity relationship optimization allowed generation of low nanomolar BRPF2 BD inhibitors with strong selectivity against BRPF1 and BRPF3 BDs. In addition, a strong inhibition of TAF1/TAF1L BD2 was measured for most derivatives. The best compound of the series was BAY-299, which is a very potent, dual inhibitor with an IC₅₀ of 67 nM for BRPF2 BD, 8 nM for TAF1 BD2, and 106 nM for TAF1L BD2. Importantly, no activity was measured for BRD4 BDs. Furthermore, cellular activity was evidenced using a BRPF2– or TAF1–histone H3.3 or H4 interaction assay.

Efficiency of different measures for defining the applicability domain of classification models

The goal of defining an applicability domain for a predictive classification model is to identify the region in chemical space where the model's predictions are reliable. The boundary of the applicability domain is defined with the help of a measure that shall reflect the reliability of an individual prediction. Here, the available measures are differentiated into those that flag unusual objects and which are independent of the original classifier and those that use information of the trained classifier. The former set of techniques is referred to as novelty detection while the latter is designated as confidence estimation. A review of the available confidence estimators shows that most of these measures estimate the probability of class membership of the predicted objects which is inversely related to the error probability. Thus, class probability estimates are natural candidates for defining the applicability domain but were not comprehensively included in previous benchmark studies. The focus of the present study is to find the best measure for defining the applicability domain for a given binary classification technique and to determine the performance of novelty detection versus confidence estimation. Six different binary classification techniques in combination with ten data sets were studied to benchmark the various measures. The area under the receiver operating characteristic curve (AUC ROC) was employed as main benchmark criterion. It is shown that class probability estimates constantly perform best to differentiate between reliable and unreliable predictions. Previously proposed alternatives to class probability estimates do not perform better than the latter and are inferior in most cases. Interestingly, the impact of defining an applicability domain depends on the observed area under the receiver operator characteristic curve. That means that it depends on the level of difficulty of the classification problem (expressed as AUC ROC) and will be largest for intermediately difficult problems (range AUC ROC 0.7-0.9). In the ranking of classifiers, classification random forests performed best on average. Hence, classification random forests in combination with the respective class probability estimate are a good starting point for predictive binary chemoinformatic classifiers with applicability domain. Graphical abstract .

Discovery and Characterization of the Potent and Selective P2X4 Inhibitor N-[4-(3-Chlorophenoxy)-3-sulfamoylphenyl]-2-phenylacetamide (BAY-1797) and Structure-Guided Amelioration of Its CYP3A4 Induction Profile

The P2X4 receptor is a ligand-gated ion channel that is expressed on a variety of cell types, especially those involved in inflammatory and immune processes. High-throughput screening led to a new class of P2X4 inhibitors with substantial CYP 3A4 induction in human hepatocytes. A structure-guided optimization with respect to decreased pregnane X receptor (PXR) binding was started. It was found that the introduction of larger and more polar substituents on the ether linker led to less PXR binding while maintaining the P2X4 inhibitory potency. This translated into significantly reduced CYP 3A4 induction for compounds 71 and 73. Unfortunately, the in vivo pharmacokinetic (PK) profiles of these compounds were insufficient for the desired profile in humans. However, BAY-1797 (10) was identified and characterized as a potent and selective P2X4 antagonist. This compound is suitable for in vivo studies in rodents, and the anti-inflammatory and anti-nociceptive effects of BAY-1797 were demonstrated in a mouse complete Freund's adjuvant (CFA) inflammatory pain model.

Estimating the domain of applicability for machine learning QSAR models: a study on aqueous solubility of drug discovery molecules

We investigate the use of different Machine Learning methods to construct models for aqueous solubility. Models are based on about 4000 compounds, including an in-house set of 632 drug discovery molecules of Bayer Schering Pharma. For each method, we also consider an appropriate method to obtain error bars, in order to estimate the domain of applicability (DOA) for each model. Here, we investigate error bars from a Bayesian model (Gaussian Process (GP)), an ensemble based approach (Random Forest), and approaches based on the Mahalanobis distance to training data (for Support Vector Machine and Ridge Regression models). We evaluate all approaches in terms of their prediction accuracy (in cross-validation, and on an external validation set of 536 molecules) and in how far the individual error bars can faithfully represent the actual prediction error.

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.

Machine Learning Models for Lipophilicity and Their Domain of Applicability

Unfavorable lipophilicity and water solubility cause many drug failures; therefore these properties have to be taken into account early on in lead discovery. Commercial tools for predicting lipophilicity usually have been trained on small and neutral molecules, and are thus often unable to accurately predict in-house data. Using a modern Bayesian machine learning algorithm--a Gaussian process model--this study constructs a log D7 model based on 14,556 drug discovery compounds of Bayer Schering Pharma. Performance is compared with support vector machines, decision trees, ridge regression, and four commercial tools. In a blind test on 7013 new measurements from the last months (including compounds from new projects) 81% were predicted correctly within 1 log unit, compared to only 44% achieved by commercial software. Additional evaluations using public data are presented. We consider error bars for each method (model based error bars, ensemble based, and distance based approaches), and investigate how well they quantify the domain of applicability of each model.

Mechanistic Reactivity Descriptors for the Prediction of Ames Mutagenicity of Primary Aromatic Amines

Pharmaceutical products are often synthesized by the use of reactive starting materials and intermediates. These can, either as impurities or through metabolic activation, bind to the DNA. Primary aromatic amines belong to the critical classes that are considered potentially mutagenic in the Ames test, so there is a great need for good prediction models for risk assessment. How primary aromatic amines exert their mutagenic potential can be rationalized by the widely accepted nitrenium ion hypothesis of covalent binding to the DNA of reactive electrophiles formed out of the aromatic amines. Since the reactive chemical species is different in chemical structure from the actual compound, it is difficult to achieve good predictions via classical descriptor or fingerprint-based machine learning. In this approach, we use a combination of different molecular and atomic descriptors that is able to describe different mechanistic aspects of the metabolic transformation leading from the primary aromatic amine to the reactive metabolite that binds to the DNA. Applied to a test set, the combination shows significantly better performance than models that only use one of these descriptors and complemented the general internal Ames mutagenicity prediction model at Bayer.

Discovery and Characterization of the Potent and Highly Selective 1,7-Naphthyridine-Based Inhibitors BAY-091 and BAY-297 of the Kinase PIP4K2A

PIP4K2A is an insufficiently studied type II lipid kinase that catalyzes the conversion of phosphatidylinositol-5-phosphate (PI5P) into phosphatidylinositol 4,5-bisphosphate (PI4,5P₂). The involvement of PIP4K2A/B in cancer has been suggested, particularly in the context of p53 mutant/null tumors. PIP4K2A/B depletion has been shown to induce tumor growth inhibition, possibly due to hyperactivation of AKT and reactive oxygen species-mediated apoptosis. Herein, we report the identification of the novel potent and highly selective inhibitors BAY-091 and BAY-297 of the kinase PIP4K2A by high-throughput screening and subsequent structure-based optimization. Cellular target engagement of BAY-091 and BAY-297 was demonstrated using cellular thermal shift assay technology. However, inhibition of PIP4K2A with BAY-091 or BAY-297 did not translate into the hypothesized mode of action and antiproliferative activity in p53-deficient tumor cells. Therefore, BAY-091 and BAY-297 serve as valuable chemical probes to study PIP4K2A signaling and its involvement in pathophysiological conditions such as cancer.

Discovery and optimization of pyridyl-cycloalkyl-carboxylic acids as inhibitors of microsomal prostaglandin E synthase-1 for the treatment of endometriosis

Here we report on novel and potent pyridyl-cycloalkyl-carboxylic acid inhibitors of microsomal prostaglandin E synthase-1 (PTGES). PTGES produces, as part of the prostaglandin pathway, prostaglandin E2 which is a well-known driver for pain and inflammation. This fact together with the observed upregulation of PTGES during inflammation suggests that blockade of the enzyme might provide a beneficial treatment option for inflammation related conditions such as endometriosis. Compound 5a, a close analogue of the screening hit, potently inhibited PTGES in vitro, displayed excellent PK properties in vitro and in vivo and demonstrated efficacy in a CFA-induced pain model in mice and in a rat dyspareunia endometriosis model and was therefore selected for further studies.

Large-scale evaluation of k-fold cross-validation ensembles for uncertainty estimation

It is insightful to report an estimator that describes how certain a model is in a prediction, additionally to the prediction alone. For regression tasks, most approaches implement a variation of the ensemble method, apart from few exceptions. Instead of a single estimator, a group of estimators yields several predictions for an input. The uncertainty can then be quantified by measuring the disagreement between the predictions, for example by the standard deviation. In theory, ensembles should not only provide uncertainties, they also boost the predictive performance by reducing errors arising from variance. Despite the development of novel methods, they are still considered the "golden-standard" to quantify the uncertainty of regression models. Subsampling-based methods to obtain ensembles can be applied to all models, regardless whether they are related to deep learning or traditional machine learning. However, little attention has been given to the question whether the ensemble method is applicable to virtually all scenarios occurring in the field of cheminformatics. In a widespread and diversified attempt, ensembles are evaluated for 32 datasets of different sizes and modeling difficulty, ranging from physicochemical properties to biological activities. For increasing ensemble sizes with up to 200 members, the predictive performance as well as the applicability as uncertainty estimator are shown for all combinations of five modeling techniques and four molecular featurizations. Useful recommendations were derived for practitioners regarding the success and minimum size of ensembles, depending on whether predictive performance or uncertainty quantification is of more importance for the task at hand.

Best of both worlds: An expansion of the state of the art pKa model with data from three industrial partners

In a unique collaboration between Simulations Plus and several industrial partners, we were able to develop a new version 11.0 of the previously published in silico pKa model, S+pKa, with considerably improved prediction accuracy. The model's training set was vastly expanded by large amounts of experimental data obtained from F. Hoffmann-La Roche AG, Genentech Inc., and the Crop Science division of Bayer AG. The previous v7.0 of S+pKa was trained on data from public sources and the Pharmaceutical division of Bayer AG. The model has shown dramatic improvements in predictive accuracy when externally validated on three new contributor compound sets. Less expected was v11.0's improvement in prediction on new compounds developed at Bayer Pharma after v7.0 was released (2013-2023), even without contributing additional data to v11.0. We illustrate chemical space coverage by chemistries encountered in the five domains, public and industrial, outline model construction, and discuss factors contributing to model's success.

Abstract 980: BAY-299, a novel chemical probe for in-depth analysis of the function of the bromodomain proteins BRPF2 and TAF1

With the exception of the bromodomain and extra-terminal domain BET subgroup, little is known about the role of bromodomain (BD) containing proteins in cancer so that there is a dire need for chemical probes addressing other family members. The bromodomain and PHD-finger (BRPF) family encompasses three paralogs, BRPF1, BRPF2 and BRPF3, which are all found in histone acetyltransferase (HAT) complexes. BRPF2 is a scaffold protein and its knock-out leads to embryonic lethality at E15.5, potentially due to its role in embryonic stem cell differentiation. Here we present the structure-activity relationship (SAR) and characterization of the first selective BRPF2 chemical probe BAY-299, with additionnal strong activity at TAF1, a major component of the basal transcription initiation complex TFIID. BAY-299 shows in vitro activity for BRPF2 (IC50 = 67 nM) and TAF1 second bromodomain (BD2; IC50 = 8 nM) in the TR-FRET assay, as well as in the cellular NanoBRET assay [IC50 (BRPF2) = 575 nM; IC50 (TAF1 BD2) = 825 nM]. To the best of our knowledge BAY-299 is the only disclosed inhibitor showing BRPF2 selectivity over its two paralogues BRPF1 and BRPF3. It belongs to the 1,3-benzimidazolone scaffold and bears a novel substitution which is responsible for its high BRPF2 selectivity and also for its inactivity on BET BDs. The dual inhibitory properties of BAY-299 against BRPF2 and TAF1 make it an ideal research tool for further investigation of these two proteins in physiological and pathological processes.

Citation Format: Lea Bouche, Clara D. Christ, Stephan Siegel, Cynthia Tallant, Amaury E. Fernández-Montalván, Kilian V. Huber, Verra Pütter, Susanne Müller, Oleg Fedorov, Antonius ter Laak, Tatsuo Sugawara, Detlef Stöckigt, Julia Meier, Simon J. Holton, Ingo V. Hartung, Bernard Haendler. BAY-299, a novel chemical probe for in-depth analysis of the function of the bromodomain proteins BRPF2 and TAF1 [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 980. doi:10.1158/1538-7445.AM2017-980

Abstract 5239: Probing the cancer epigenome: empowering target validation by open innovation

Low reproducibility of published target validation studies as well as the frequent failure of genetic knock-down effects to phenocopy those of small molecule inhibitors have been recognized as road blocks for cancer drug discovery. Academic and industrial institutions have started to address these issues by providing access to high quality small molecular probes for novel targets of interest. Here we discuss probe discovery challenges and quality criteria based on the generation of three novel inhibitors for epigenetic targets.

ATAD2 (ATPase family AAA-domain containing protein 2) is an epigenetic regulator that binds to chromatin through its bromodomain (BD). ATAD2 has been proposed to act as a co-factor for oncogenic transcription factors such as ERα and Myc. A more thorough validation of ATAD2 as a therapeutic target has been hampered by the lack of appropriate ATAD2 inhibitors. Here we disclose a structurally unprecedented series of ATAD2 BD inhibitors identified from a DNA-encoded library screen. Optimization delivered BAY-850, a highly potent and exceptionally selective ATAD2 BD inhibitor, which fully recapitulates effects seen by genetic mutagenesis studies in a cellular assay.

The three BD and PHD-finger (BRPF) family members are found in histone acetyltransferase complexes. Whereas bromodomain inhibitors with dual activity against BRPF1 and 2 have been described before, we now disclose BAY-299, the first nanomolar inhibitor of the BRPF2 BD with high selectivity against its paralogs. Isoform selectivity was confirmed in cellular protein-protein interaction assays and rationalized based on X-Ray structures.

BAY-598, a highly selective, cellularly active and orally bioavailable inhibitor of the protein lysine methyl transferase SMYD2, had been disclosed previously (Stresemann et al., AACR 2015). Development of BAY-598 allowed the identification of new methylation targets of SMYD2 as well as a proposed role of SMYD2 in pancreatic cancer.

These results support further development of small molecule inhibitors as research tools to probe the functional role of novel epigenetic targets and underscore the power of open innovation for advancing our understanding of cancer target biology.

Citation Format: Ingo V. Hartung, Cheryl Arrowsmith, Volker Badock, Naomi Barak, Markus Berger, Peter J. Brown, Clara D. Christ, Erik Eggert, Ursula Egner, Oleg Fedorov, Amaury E. Fernandez-Montalvan, Matyas Gorjanacz, Andrea Haegebarth, Bernard Haendler, Roman C. Hillig, Simon H. Holton, Kilian V. Huber, Seong J. Koo, Antonius ter Laak, Susanne Mueller, Anke Mueller-Fahrnow, Cora Scholten, Stephan Siegel, Timo Stellfeld, Detlef Stoeckigt, Carlo Stresemann, Masoud Vedadi, Joerg Weiske, Hilmar Weinmann. Probing the cancer epigenome: empowering target validation by open innovation [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5239. doi:10.1158/1538-7445.AM2017-5239

Discovery and Characterization of BAY-184: A New Potent and Selective Acylsulfonamide-Benzofuran In Vivo-Active KAT6AB Inhibitor

KAT6A and KAT6B genes are two closely related lysine acetyltransferases that transfer an acetyl group from acetyl coenzyme A (AcCoA) to lysine residues of target histone substrates, hence playing a key role in chromatin regulation. KAT6A and KAT6B genes are frequently amplified in various cancer types. In breast cancer, the 8p11-p12 amplicon occurs in 12-15% of cases, resulting in elevated copy numbers and expression levels of chromatin modifiers like KAT6A. Here, we report the discovery of a new acylsulfonamide-benzofuran series as a novel structural class for KAT6A/B inhibition. These compounds were identified through high-throughput screening and subsequently optimized using molecular modeling and cocrystal structure determination. The final tool compound, BAY-184 (29), was successfully validated in an in vivo proof-of-concept study.