In silico systems for predicting chemical-induced side effects using known and potential chemical protein interactions, enabling mechanism estimation

Pathway-based prediction of the therapeutic effects and mode of action of custom-made multiherbal medicines

Multiherbal medicines are traditionally used as personalized medicines with custom combinations of crude drugs; however, the mechanisms of multiherbal medicines are unclear. In this study, we developed a novel pathway-based method to predict therapeutic effects and the mode of action of custom-made multiherbal medicines using machine learning. This method considers disease-related pathways as therapeutic targets and evaluates the comprehensive influence of constituent compounds on their potential target proteins in the disease-related pathways. Our proposed method enabled us to comprehensively predict new indications of 194 Kampo medicines for 87 diseases. Using Kampo-induced transcriptomic data, we demonstrated that Kampo constituent compounds stimulated the disease-related proteins and a customized Kampo formula enhanced the efficacy compared with an existing Kampo formula. The proposed method will be useful for discovering effective Kampo medicines and optimizing custom-made multiherbal medicines in practice.

Revealing Comprehensive Food Functionalities and Mechanisms of Action through Machine Learning

Foods possess a range of unexplored functionalities; however, fully identifying these functions through empirical means presents significant challenges. In this study, we have proposed an in silico approach to comprehensively predict the functionalities of foods, encompassing even processed foods. This prediction is accomplished through the utilization of machine learning on biomedical big data. Our focus revolves around disease-related protein pathways, wherein we statistically evaluate how the constituent compounds collaboratively regulate these pathways. The proposed method has been employed across 876 foods and 83 diseases, leading to an extensive revelation of both food functionalities and their underlying operational mechanisms. Additionally, this approach identifies food combinations that potentially affect molecular pathways based on interrelationships between food functions within disease-related pathways. Our proposed method holds potential for advancing preventive healthcare.

Predicting therapeutic and side effects from drug binding affinities to human proteome structures

Evaluation of the binding affinities of drugs to proteins is a crucial process for identifying drug pharmacological actions, but it requires three dimensional structures of proteins. Herein, we propose novel computational methods to predict the therapeutic indications and side effects of drug candidate compounds from the binding affinities to human protein structures on a proteome-wide scale. Large-scale docking simulations were performed for 7,582 drugs with 19,135 protein structures revealed by AlphaFold (including experimentally unresolved proteins), and machine learning models on the proteome-wide binding affinity score (PBAS) profiles were constructed. We demonstrated the usefulness of the method for predicting the therapeutic indications for 559 diseases and side effects for 285 toxicities. The method enabled to predict drug indications for which the related protein structures had not been experimentally determined and to successfully extract proteins eliciting the side effects. The proposed method will be useful in various applications in drug discovery.

RAID: Regression Analysis–Based Inductive DNA Microarray for Precise Read-Across

Chemical structure-based read-across represents a promising method for chemical toxicity evaluation without the need for animal testing; however, a chemical structure is not necessarily related to toxicity. Therefore, in vitro studies were often used for read-across reliability refinement; however, their external validity has been hindered by the gap between in vitro and in vivo conditions. Thus, we developed a virtual DNA microarray, regression analysis–based inductive DNA microarray (RAID), which quantitatively predicts in vivo gene expression profiles based on the chemical structure and/or in vitro transcriptome data. For each gene, elastic-net models were constructed using chemical descriptors and in vitro transcriptome data to predict in vivo data from in vitro data (in vitro to in vivo extrapolation; IVIVE). In feature selection, useful genes for assessing the quantitative structure–activity relationship (QSAR) and IVIVE were identified. Predicted transcriptome data derived from the RAID system reflected the in vivo gene expression profiles of characteristic hepatotoxic substances. Moreover, gene ontology and pathway analysis indicated that nuclear receptor-mediated xenobiotic response and metabolic activation are related to these gene expressions. The identified IVIVE-related genes were associated with fatty acid, xenobiotic, and drug metabolisms, indicating that in vitro studies were effective in evaluating these key events. Furthermore, validation studies revealed that chemical substances associated with these key events could be detected as hepatotoxic biosimilar substances. These results indicated that the RAID system could represent an alternative screening test for a repeated-dose toxicity test and toxicogenomics analyses. Our technology provides a critical solution for IVIVE-based read-across by considering the mode of action and chemical structures.

Development of in silico prediction models for drug-induced liver malignant tumors based on the activity of molecular initiating events: Biologically interpretable features

Liver malignant tumors (LMTs) have recently been reported as severe and life-threatening adverse drug events associated with drug-induced liver injury (DILI). DILIs are the most common adverse drug event and can cause the withdrawal of medicinal products or major regulatory action. To reduce the attrition rate and cost of drug discovery, various quantitative structure-toxicity relationship models have been proposed to predict the probability of a DILI based on the chemical structure of a drug. However, there are many unresolved issues regarding the predictors of LMT-inducing drugs, and biologically interpretable prediction models for LMT have not been developed. Here, we constructed prediction models for whether a drug is LMT-inducing based on the activity of molecular initiating events (MIEs), which are biologically interpretable features and are defined as the initial interaction between a molecule and biosystem. We then constructed five machine learning models (i.e., LightGBM, XGBoost, random forest, neural network, and support vector machine) and evaluated their predictive performances. LightGBM achieved the best performance among the tested models. The MIEs making the highest contribution to the model construction for drug-induced LMT were inducement of Enhanced Level of Genome Instability Gene 1 (human ATAD5), nuclear factor-κ B, and activation of thyrotropin-releasing hormone receptor. These results support the previous literature and can be related to the mechanism onset of drug-induced LMT. Our findings may provide useful knowledge for drug development, research, and regulatory decision-making and will contribute to building more accurate and meaningful DILI prediction models by increasing understanding of biological predictors.

TRIOMPHE: Transcriptome-Based Inference and Generation of Molecules with Desired Phenotypes by Machine Learning

One of the most challenging tasks in the drug-discovery process is the efficient identification of small molecules with desired phenotypes. In this study, we propose a novel computational method for omics-based de novo drug design, which we call TRIOMPHE (transcriptome-based inference and generation of molecules with desired phenotypes). We investigated the correlation between chemically induced transcriptome profiles (reflecting cellular responses to compound treatment) and genetically perturbed transcriptome profiles (reflecting cellular responses to gene knock-down or gene overexpression of target proteins) in terms of ligand-target interactions. Subsequently, we developed novel machine learning methods to generate the chemical structures of new molecules with desired transcriptome profiles in the framework of a variational autoencoder. The use of desired transcriptome profiles enables the automatic design of molecules that are likely to have bioactivities for target proteins of interest. We showed that our methods can generate chemically valid molecules that are likely to have biological activities on 10 target proteins; moreover, they can outperform previous methods that had the same objective. Our omics-based structure generator is expected to be useful for the de novo design of drugs for a variety of target proteins.

In Silico Model for Chemical-Induced Chromosomal Damages Elucidates Mode of Action and Irrelevant Positives

In silico tools to predict genotoxicity have become important for high-throughput screening of chemical substances. However, current in silico tools to evaluate chromosomal damage do not discriminate in vitro-specific positives that can be followed by in vivo tests. Herein, we establish an in silico model for chromosomal damages with the following approaches: (1) re-categorizing a previous data set into three groups (positives, negatives, and misleading positives) according to current reports that use weight-of-evidence approaches and expert judgments; (2) utilizing a generalized linear model (Elastic Net) that uses partial structures of chemicals (organic functional groups) as explanatory variables of the statistical model; and (3) interpreting mode of action in terms of chemical structures identified. The accuracy of our model was 85.6%, 80.3%, and 87.9% for positive, negative, and misleading positive predictions, respectively. Selected organic functional groups in the models for positive prediction were reported to induce genotoxicity via various modes of actions (e.g., DNA adduct formation), whereas those for misleading positives were not clearly related to genotoxicity (e.g., low pH, cytotoxicity induction). Therefore, the present model may contribute to high-throughput screening in material design or drug discovery to verify the relevance of estimated positives considering their mechanisms of action.