The application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale... Show moreThe application of multiple omics technologies in biomedical cohorts has the potential to reveal patient-level disease characteristics and individualized response to treatment. However, the scale and heterogeneous nature of multi-modal data makes integration and inference a non-trivial task. We developed a deep-learning-based framework, multi-omics variational autoencoders (MOVE), to integrate such data and applied it to a cohort of 789 people with newly diagnosed type 2 diabetes with deep multi-omics phenotyping from the DIRECT consortium. Using in silico perturbations, we identified drug-omics associations across the multi-modal datasets for the 20 most prevalent drugs given to people with type 2 diabetes with substantially higher sensitivity than univariate statistical tests. From these, we among others, identified novel associations between metformin and the gut microbiota as well as opposite molecular responses for the two statins, simvastatin and atorvastatin. We used the associations to quantify drug-drug similarities, assess the degree of polypharmacy and conclude that drug effects are distributed across the multi-omics modalities. Show less
This read-across case study characterises thirteen, structurally similar carboxylic acids demonstrating the application of in vitro and in silico human-based new approach methods, to determine... Show moreThis read-across case study characterises thirteen, structurally similar carboxylic acids demonstrating the application of in vitro and in silico human-based new approach methods, to determine biological similarity. Based on data from in vivo animal studies, the read-across hypothesis is that all analogues are steatotic and so should be considered hazardous. Transcriptomic analysis to determine differentially expressed genes (DEGs) in hepatocytes served as first tier testing to confirm a common mode-of-action and identify differences in the potency of the analogues. An adverse outcome pathway (AOP) network for hepatic steatosis, informed the design of an in vitro testing battery, targeting AOP relevant MIEs and KEs, and Dempster-Shafer decision theory was used to systematically quantify uncertainty and to define the minimal testing scope. The case study shows that the read-across hypothesis is the critical core to designing a robust, NAM-based testing strategy. By summarising the current mechanistic understanding, an AOP enables the selection of NAMs covering MIEs, early KEs, and late KEs. Experimental coverage of the AOP in this way is vital since MIEs and early KEs alone are not confirmatory of progression to the AO. This strategy exemplifies the workflow previously published by the EUTOXRISK project driving a paradigm shift towards NAM-based NGRA. Show less
Drug-induced liver injury (DILI) cannot be accurately predicted by animal models. In addition, currently available in vitro methods do not allow for the estimation of hepatotoxic doses or the... Show moreDrug-induced liver injury (DILI) cannot be accurately predicted by animal models. In addition, currently available in vitro methods do not allow for the estimation of hepatotoxic doses or the determination of an acceptable daily intake (ADI). To overcome this limitation, an in vitro/in silico method was established that predicts the risk of human DILI in relation to oral doses and blood concentrations. This method can be used to estimate DILI risk if the maximal blood concentration (Cmax) of the test compound is known. Moreover, an ADI can be estimated even for compounds without information on blood concentrations. To systematically optimize the in vitro system, two novel test performance metrics were introduced, the toxicity separation index (TSI) which quantifies how well a test differentiates between hepatotoxic and non-hepatotoxic compounds, and the toxicity estimation index (TEI) which measures how well hepatotoxic blood concentrations in vivo can be estimated. In vitro test performance was optimized for a training set of 28 compounds, based on TSI and TEI, demonstrating that (1) concentrations where cytotoxicity first becomes evident in vitro (EC10) yielded better metrics than higher toxicity thresholds (EC50); (2) compound incubation for 48 h was better than 24 h, with no further improvement of TSI after 7 days incubation; (3) metrics were moderately improved by adding gene expression to the test battery; (4) evaluation of harmacokinetic parameters demonstrated that total blood compound concentrations and the 95%-population-based percentile of Cmax were best suited to estimate human toxicity. With a support vector machine-based classifier, using EC10 and Cmax as variables, the cross-validated sensitivity, specificity and accuracy for hepatotoxicity prediction were 100, 88 and 93%, respectively. Concentrations in the culture medium allowed extrapolation to blood concentrations in vivo that are associated with a specific probability of hepatotoxicity and the corresponding oral doses were obtained by reverse modeling. Application of this in vitro/in silico method to the rat hepatotoxicant pulegone resulted in an ADI that was similar to values previously established based on animal experiments. In conclusion, the proposed method links oral doses and blood concentrations of test compounds to the probability of hepatotoxicity. Show less
