In silico toxicology is an emerging field. It gains increasing importance as research is aiming to decrease the use of animal experiments as suggested in the 3R principles by Russell and Burch. In silico toxicology is a means to identify hazards of compounds before synthesis, and thus in very early stages of drug development. For chemical industries, as well as regulatory agencies it can aid in gap‐filling and guide risk minimization strategies. Techniques such as structural alerts, read‐across, quantitative structure–activity relationship, machine learning, and deep learning allow to use in silico toxicology in many cases, some even when data is scarce. Especially the concept of adverse outcome pathways puts all techniques into a broader context and can elucidate predictions by mechanistic insights. This article is categorized under: Structure and Mechanism > Computational Biochemistry and Biophysics Data Science > Chemoinformatics

Solute Carriers (SLCs) represent the largest family of transmembrane transporters in humans and constitute major determinants of cellular metabolism. Several SLCs have been shown to be required for the uptake of chemical compounds into cellular systems, but systematic surveys of transporter-drug relationships in human cells are currently lacking. We performed a series of genetic screens in a haploid human cell line against 60 cytotoxic compounds representative of the chemical space populated by approved drugs. By using an SLC-focused CRISPR/Cas9 library, we identified transporters whose absence induced resistance to the drugs tested. This included dependencies involving the transporters SLC11A2/SLC16A1 for artemisinin derivatives and SLC35A2/SLC38A5 for cisplatin. The functional dependence on SLCs observed for a significant proportion of the compounds screened suggests a widespread role for SLCs in the uptake and cellular activity of cytotoxic drugs and provides an experimentally validated set of SLC-drug associations for a number of clinically relevant compounds.

Over the last few years more and more organ and idiosyncratic toxicities were linked to mitochondrial toxicity. Despite well-established assays, such as the seahorse and Glucose/Galactose assay, an in silico approach to mitochondrial toxicity is still feasible, particularly when it comes to the assessment of large compound libraries. Therefore, in silico approaches could be very beneficial to indicate hazards early in the drug development pipeline. By combining multiple endpoints, we derived the largest so far published dataset on mitochondrial toxicity. A thorough data analysis shows that molecules causing mitochondrial toxicity can be distinguished by physicochemical properties. Finally, the combination of machine learning and structural alerts highlights the suitability for in silico risk assessment of mitochondrial toxicity.

The activity and potency of a drug is inherently affected by the metabolic state of its target cell.Solute Carriers (SLCs) represent the largest family of transmembrane transporters in humans and constitute major determinants of cellular metabolism. Several SLCs have been shown to be required for the uptake of individual chemical compounds into cellular systems, but systematic surveys of transporter-drug relationships in human cells are currently lacking. We performed a series of genetic screens in the haploid human cell line HAP1 using a set of 60 cytotoxic compounds representative of the chemical space populated by approved drugs. By using a SLC-focused CRISPR/Cas9 lentiviral library, we identified transporters whose absence induced resistance to the drugs tested. Among the hundreds of drug-SLC relationships identified, we confirmed the role of the folate transporter SLC19A1 on the activity of antifolates and of SLC29A1 on several nucleoside analogs. Among the newly discovered dependencies, we identified the transporters SLC11A2/SLC16A1 for artemisinin derivatives and SLC35A2/SLC38A5 for cisplatin. The functional dependence on SLCs observed for a significant proportion of the compounds screened suggested a widespread role for SLCs in the uptake and cellular activity of cytotoxic drugs and provided an experimentally validated set of SLC-drug associations for a number of clinically relevant compounds.

Training neural networks with small and imbalanced datasets often leads to overfitting and disregard of the minority class. For predictive toxicology, however, models with a good balance between sensitivity and specificity are needed. In this paper we introduce conformational oversampling as a means to balance and oversample datasets for prediction of toxicity. Conformational oversampling enhances a dataset by generation of multiple conformations of a molecule. These conformations can be used to balance, as well as oversample a dataset, thereby increasing the dataset size without the need of artificial samples. We show that conformational oversampling facilitates training of neural networks and provides state-of-the-art results on the Tox21 dataset.