A Symbolic Regression Model for the Prediction of Drug Binding to Human Liver Microsomes.

It is common practice in the early drug discovery process to conduct in vitro screening experiments using liver microsomes in order to obtain an initial assessment of test compound metabolic stability. Compounds which bind to liver microsomes are unavailable for interaction with the drug metabolizing enzymes. As such, assessment of the unbound fraction of compound available for biotransformation is an important factor for interpretation of in vitro experimental results and to improve prediction of the in vivo metabolic clearance. Various in silico methods have been proposed for the prediction of test compound binding to microsomes, from various simple lipophilicity-based models with moderate performance to sophisticated machine learning models which demonstrate superior performance at the cost of increased complexity and higher data requirements. In this work, we attempt to strike a middle ground by developing easily implementable equations with improved predictive performance. We employ a symbolic regression approach based on a medium-size in-house data set of fraction unbound in human liver microsomes measurements allowing the identification of novel equations with improved performance. We validate the model performance on an in-house held-out test set and an external validation set.

Bidirectional Graphormer for Reactivity Understanding: Neural Network Trained to Reaction Atom-to-Atom Mapping Task.

This work introduces GraphormerMapper, a new algorithm for reaction atom-to-atom mapping (AAM) based on a transformer neural network adopted for the direct processing of molecular graphs as sets of atoms and bonds, as opposed to SMILES/SELFIES sequence-based approaches, in combination with the Bidirectional Encoder Representations from Transformers (BERT) network. The graph transformer serves to extract molecular features that are tied to atoms and bonds. The BERT network is used for chemical transformation learning. In a benchmarking study with IBM RxnMapper, which is the best AAM algorithm according to our previous study, we demonstrate that our AAM algorithm is superior to it on our "Golden" benchmarking data set.

Dataset for quantum-mechanical exploration of conformers and solvent effects in large drug-like molecules.

We here introduce the Aquamarine (AQM) dataset, an extensive quantum-mechanical (QM) dataset that contains the structural and electronic information of 59,783 low-and high-energy conformers of 1,653 molecules with a total number of atoms ranging from 2 to 92 (mean: 50.9), and containing up to 54 (mean: 28.2) non-hydrogen atoms. To gain insights into the solvent effects as well as collective dispersion interactions for drug-like molecules, we have performed QM calculations supplemented with a treatment of many-body dispersion (MBD) interactions of structures and properties in the gas phase and implicit water. Thus, AQM contains over 40 global and local physicochemical properties (including ground-state and response properties) per conformer computed at the tightly converged PBE0+MBD level of theory for gas-phase molecules, whereas PBE0+MBD with the modified Poisson-Boltzmann (MPB) model of water was used for solvated molecules. By addressing both molecule-solvent and dispersion interactions, AQM dataset can serve as a challenging benchmark for state-of-the-art machine learning methods for property modeling and de novo generation of large (solvated) molecules with pharmaceutical and biological relevance.

Atom-to-atom Mapping: A Benchmarking Study of Popular Mapping Algorithms and Consensus Strategies.

In this paper, we compare the most popular Atom-to-Atom Mapping (AAM) tools: ChemAxon,[1] Indigo,[2] RDTool,[3] NameRXN (NextMove),[4] and RXNMapper[5] which implement different AAM algorithms. An open-source RDTool program was optimized, and its modified version ("new RDTool") was considered together with several consensus mapping strategies. The Condensed Graph of Reaction approach was used to calculate chemical distances and develop the "AAM fixer" algorithm for an automatized correction of erroneous mapping. The benchmarking calculations were performed on a Golden dataset containing 1851 manually mapped and curated reactions. The best performing RXNMapper program together with the AMM Fixer was applied to map the USPTO database. The Golden dataset, mapped USPTO and optimized RDTool are available in the GitHub repository https://github.com/Laboratoire-de-Chemoinformatique.

Reaction Data Curation I: Chemical Structures and Transformations Standardization.

The quality of experimental data for chemical reactions is a critical consideration for any reaction-driven study. However, the curation of reaction data has not been extensively discussed in the literature so far. Here, we suggest a 4 steps protocol that includes the curation of individual structures (reactants and products), chemical transformations, reaction conditions and endpoints. Its implementation in Python3 using CGRTools toolkit has been used to clean three popular reaction databases Reaxys, USPTO and Pistachio. The curated USPTO database is available in the GitHub repository (Laboratoire-de-Chemoinformatique/Reaction_Data_Cleaning).

Industry-scale application and evaluation of deep learning for drug target prediction.

Artificial intelligence (AI) is undergoing a revolution thanks to the breakthroughs of machine learning algorithms in computer vision, speech recognition, natural language processing and generative modelling. Recent works on publicly available pharmaceutical data showed that AI methods are highly promising for Drug Target prediction. However, the quality of public data might be different than that of industry data due to different labs reporting measurements, different measurement techniques, fewer samples and less diverse and specialized assays. As part of a European funded project (ExCAPE), that brought together expertise from pharmaceutical industry, machine learning, and high-performance computing, we investigated how well machine learning models obtained from public data can be transferred to internal pharmaceutical industry data. Our results show that machine learning models trained on public data can indeed maintain their predictive power to a large degree when applied to industry data. Moreover, we observed that deep learning derived machine learning models outperformed comparable models, which were trained by other machine learning algorithms, when applied to internal pharmaceutical company datasets. To our knowledge, this is the first large-scale study evaluating the potential of machine learning and especially deep learning directly at the level of industry-scale settings and moreover investigating the transferability of publicly learned target prediction models towards industrial bioactivity prediction pipelines.

High-Throughput Gene Expression Profiles to Define Drug Similarity and Predict Compound Activity.

By adding biological information, beyond the chemical properties and desired effect of a compound, uncharted compound areas and connections can be explored. In this study, we add transcriptional information for 31K compounds of Janssen's primary screening deck, using the HT L1000 platform and assess (a) the transcriptional connection score for generating compound similarities, (b) machine learning algorithms for generating target activity predictions, and (c) the scaffold hopping potential of the resulting hits. We demonstrate that the transcriptional connection score is best computed from the significant genes only and should be interpreted within its confidence interval for which we provide the stats. These guidelines help to reduce noise, increase reproducibility, and enable the separation of specific and promiscuous compounds. The added value of machine learning is demonstrated for the NR3C1 and HSP90 targets. Support Vector Machine models yielded balanced accuracy values ≥80% when the expression values from DDIT4 & SERPINE1 and TMEM97 & SPR were used to predict the NR3C1 and HSP90 activity, respectively. Combining both models resulted in 22 new and confirmed HSP90-independent NR3C1 inhibitors, providing two scaffolds (i.e., pyrimidine and pyrazolo-pyrimidine), which could potentially be of interest in the treatment of depression (i.e., inhibiting the glucocorticoid receptor (i.e., NR3C1), while leaving its chaperone, HSP90, unaffected). As such, the initial hit rate increased by a factor 300, as less, but more specific chemistry could be screened, based on the upfront computed activity predictions.

Chemogenomics approaches for receptor deorphanization and extensions of the chemogenomics concept to phenotypic space.

Chemogenomic approaches, which link ligand chemistry to bioactivity against targets (and, by extension, to phenotypes) are becoming more and more important due to the increasing number of bioactivity data available both in proprietary databases as well as in the public domain. In this article we review chemogenomics approaches applied in four different domains: Firstly, due to the relationship between protein targets from which an approximate relation between their respective bioactive ligands can be inferred, we investigate the extent to which chemogenomics approaches can be applied to receptor deorphanization. In this case it was found that by using knowledge about active compounds of related proteins, in 93% of all cases enrichment better than random could be obtained. Secondly, we analyze different cheminformatics analysis methods with respect to their behavior in chemogenomics studies, such as subgraph mining and Bayesian models. Thirdly, we illustrate how chemogenomics, in its particular flavor of 'proteochemometrics', can be applied to extrapolate bioactivity predictions from given data points to related targets. Finally, we extend the concept of 'chemogenomics' approaches, relating ligand chemistry to bioactivity against related targets, into phenotypic space which then falls into the area of 'chemical genomics' and 'chemical genetics'; given that this is very often the desired endpoint of approaches in not only the pharmaceutical industry, but also in academic probe discovery, this is often the endpoint the experimental scientist is most interested in.