Daniel Gusenleitner wins 1st Place Biomedical Student Poster Competition
BUSRP trainee Daniel Gusenleitner won first place in the Biomedical Student Poster Competition this month at the Superfund Annual Meeting in Baton Rouge, LA. The poster entitled Rodent-based Toxicogenomic Models of Hepatocarcinogenicity is a collaborative effort between investigators of the Bioinformatics Core, Project 3, & NIEHS. To read more about Daniel & his research interests, click here.
ABSTRACT: Rodent-based Toxicogenomic Models of Hepatocarcinogenicity
Daniel Gusenleitner1, Scott Auerbach2, David Sherr3, Stefano Monti1
1Bioinformatics, Boston University 2National Institute of Environmental Health Sciences 3Environmental Health, Boston University
Despite an overall decrease in incidence and mortality, about 40% of Americans will be diagnosed with cancer in their lifetime, and around 20% will die of it. Current approaches to test carcinogenic chemicals adopt the 2-year rodent bioassay as the de-facto “gold-standard”. This assay is costly and time-consuming and, as a result, fewer than 2% of the chemicals on the market have actually been tested for carcinogenicity. However, evidence accumulated to date suggests that gene expression profiles of model organisms exposed to chemical compounds reflect underlying mechanisms of action, and these toxicogenomic models could be used in the prediction of chemical carcinogenicity of individual chemicals or mixtures.
In this study we used a rat-based microarray dataset from the NTP DrugMatrix Database, to test the ability of toxicogenomics to model carcinogenicity. We analyzed this 4442 gene-expression profiles treated with 255 well-characterized compounds, including genotoxic and non-genotoxic carcinogens. We applied machine-learning approaches to build a classifier that predicts (AUC: 0.78) the carcinogenic potential of compounds within the dataset. The classification models were validated on an independent, dataset from the Japanese Toxicogenomics Project and achieved comparable prediction accuracy. Furthermore, comparison of the prediction results with pathological items showed a sensitivity of up to 95.8%.
In conclusion, we substantially validated the toxicogenomic approach to predict carcinogenicity, and provided strong evidence that with a larger set of compounds and cellular contexts we should be able to improve the sensitivity and specificity of the predictions. We found that the prediction of carcinogenicity is tissue-dependent and the results also confirm and expand upon previous studies implicating DNA damage, peroxisome proliferator-activated receptor (PPAR) and aryl hydrocarbon receptor (AhR) signaling and regenerative pathology in the response to exposure.
October 21, 2013