| Discriminating Different Classes of Toxicants by Transcript Profiling Guido Steiner,1,2 Laura Suter,1 Franziska Boess,1 Rodolfo
Gasser,1 Maria Cristina de Vera,1 Silvio Albertini,1 and
Stefan Ruepp1 1Non-Clinical Drug Safety and 2Bioinformatics, F. Hoffmann-La
Roche Ltd., Basel, Switzerland Abstract
Male rats were treated with various model compounds or the appropriate vehicle controls. Most substances were either well-known hepatotoxicants or showed hepatotoxicity during preclinical testing. The aim of the present study was to determine if biological samples from rats treated with various compounds can be classified based on gene expression profiles. In addition to gene expression analysis using microarrays, a complete serum chemistry profile and liver and kidney histopathology were performed. We analyzed hepatic gene expression profiles using a supervised learning method (support vector machines ; SVMs) to generate classification rules and combined this with recursive feature elimination to improve classification performance and to identify a compact subset of probe sets with potential use as biomarkers. Two different SVM algorithms were tested, and the models obtained were validated with a compound-based external cross-validation approach. Our predictive models were able to discriminate between hepatotoxic and nonhepatotoxic compounds. Furthermore, they predicted the correct class of hepatotoxicant in most cases. We provide an example showing that a predictive model built on transcript profiles from one rat strain can successfully classify profiles from another rat strain. In addition, we demonstrate that the predictive models identify nonresponders and are able to discriminate between gene changes related to pharmacology and toxicity. This work confirms the hypothesis that compound classification based on gene expression data is feasible. Key words: liver, microarray, predictive toxicology, rat, support vector machines, toxicogenomics. Environ Health Perspect 112:1236-1248 (2004) . doi:10.1289/txg.7036 available via http://dx.doi.org/ [Online 1 July 2004]
Address correspondence to S. Ruepp, F. Hoffmann-La Roche Ltd., PRBN-S (90/5.18) , CH-4070 Basel, Switzerland. Telephone: 41 61 688 3315. Fax: 41 61 688 8101. E-mail: stefan.ruepp@roche.com Supplemental data is available online (http://ehp.niehs.nih.gov/txg/members/2004/
7036/7036supplement.pdf) . We thank M. Haiker, N. Flint, S. Romer, K. Rupp, K. Schad, and C. Zihlmann for their excellent technical support and the General Toxicology group for their support. We are also deeply indebted to C. Broger, M. Neeb, B. Gaisser, and D. Wolf from the Bioinformatics group for their excellent support. The authors declare they have no competing financial interests. Received 17 February 2004 ; accepted 1 July 2004.
The full version of this article is available for free in HTML or PDF formats. |
