Cheminformatics Analysis of Assertions Mined from Literature That Describe Drug-Induced Liver Injury in Different Species

Abstract

Drug Induced Liver Injury (DILI) is one of the main causes of drug attrition. The ability to predict the liver effects of drug candidates from their chemical structure is critical to help guiding experimental drug discovery projects towards safer medicines. In this study, we have compiled a dataset of 951 compounds reported to produce a wide range of effects in the liver in different species, comprising humans, rodents, and non-rodents. The liver effects for this dataset were obtained as assertional meta-data, generated from MEDLINE abstracts using a unique combination of lexical and linguistic methods and ontological rules. We have analyzed this dataset using conventional cheminformatics approaches and addressed several questions pertaining to cross-species concordance of liver effects, chemical determinants of liver effects in humans, and the prediction of whether a given compound is likely to cause a liver effect in humans. We found that the concordance of liver effects was relatively low (ca. 39–44%) between different species raising the possibility that species specificity could depend on specific features of chemical structure. Compounds were clustered by their chemical similarity, and similar compounds were examined for the expected similarity of their species-dependent liver effect profiles. In most cases, similar profiles were observed for members of the same cluster, but some compounds appeared as outliers. The outliers were the subject of focused assertion re-generation from MEDLINE, as well as other data sources. In some cases, additional biological assertions were identified which were in line with expectations based on compounds' chemical similarity. The assertions were further converted to binary annotations of underlying chemicals (i.e., liver effect vs. no liver effect), and binary QSAR models were generated to predict whether a compound would be expected to produce liver effects in humans. Despite the apparent heterogeneity of data, models have shown good predictive power assessed by external five-fold cross validation procedures. The external predictive power of binary QSAR models was further confirmed by their application to compounds that were retrieved or studied after the model was developed. To the best of our knowledge, this is the first study for chemical toxicity prediction that applied QSAR modeling and other cheminformatics techniques to observational data generated by the means of automated text mining with limited manual curation, opening up new opportunities for generating and modeling chemical toxicology data.

A Predictive Ligand-Based Bayesian Model for Human Drug-Induced Liver Injury

“…There have been efforts recently to use computational models to predict DILI or idiosyncratic hepatotoxicity. We are aware of at least three studies that tackled predicting DILI dmd.aspetjournals.org using either linear discriminant analysis, artificial neural networks, and machine learning algorithms (OneR) (Cruz-Monteagudo et al, 2008), support vector machine (Fourches et al, 2010a), or structural alerts (Greene et al, 2010). A major limitation of these previous global models for DILI (and for many computational toxicology models) is their use of very small test sets in all cases.…”

“…Assembling high-quality datasets for the purpose of computational analysis can be very challenging. Commonly public data sources are used as trusted resources of information and without further validation, and, as has been demonstrated or suggested in a number of previous studies, this is not appropriate (Williams et al, 2009;Fourches et al, 2010a and references therein). The set of validated chemical structures used as the training and test data were assembled from the ChemSpider database (http:// www.chemspider.com).…”

Best Practices for QSAR Model Development, Validation, and Exploitation

“…Our recent studies provide specific examples demonstrating that the use of cheminformatics approaches helped spotting gaps or errors in biological annotations of toxic compounds. [20,31] …”

Animal Models in Drug Development

“…[93] Fourches et al evaluated animal human data for 1061 compounds known to cause hepatotoxicity in humans and found that the concordance or sensitivity among species was around 39-44%. [94] The positive and negative predictive values could not be calculated from the article but would be well below 0.39. Smith and Caldwell studied twenty-three chemicals and discovered that only four were metabolized the same in humans and rats.…”