Identification of biomarkers that distinguish chemical contaminants based on gene expression profiles

Abstract

BackgroundHigh throughput transcriptomics profiles such as those generated using microarrays have been useful in identifying biomarkers for different classification and toxicity prediction purposes. Here, we investigated the use of microarrays to predict chemical toxicants and their possible mechanisms of action.ResultsIn this study, in vitro cultures of primary rat hepatocytes were exposed to 105 chemicals and vehicle controls, representing 14 compound classes. We comprehensively compared various normalization of gene expression profiles, feature selection and classification algorithms for the classification of these 105 chemicals into14 compound classes. We found that normalization had little effect on the averaged classification accuracy. Two support vector machine (SVM) methods, LibSVM and sequential minimal optimization, had better classification performance than other methods. SVM recursive feature selection (SVM-RFE) had the highest overfitting rate when an independent dataset was used for a prediction. Therefore, we developed a new feature selection algorithm called gradient method that had a relatively high training classification as well as prediction accuracy with the lowest overfitting rate of the methods tested. Analysis of biomarkers that distinguished the 14 classes of compounds identified a group of genes principally involved in cell cycle function that were significantly downregulated by metal and inflammatory compounds, but were induced by anti-microbial, cancer related drugs, pesticides, and PXR mediators.ConclusionsOur results indicate that using microarrays and a supervised machine learning approach to predict chemical toxicants, their potential toxicity and mechanisms of action is practical and efficient. Choosing the right feature and classification algorithms for this multiple category classification and prediction is critical.

Crosstalk between Receptor and Non-receptor Mediated Chemical Modes of Action in Rat Livers Converges through a Dysregulated Gene Expression Network at Tumor Suppressor Tp53

“…Several efforts have been undertaken to identify gene expression signatures in response to toxicant exposures and to classify chemicals according to their molecular fingerprints (Amin et al, 2002 ; Bushel et al, 2002 ; Hamadeh et al, 2002a , b ; Kleinjans, 2014 ; Wei et al, 2014 ). There are several known cases of chemicals that exert their effect through a particular MOA and have overlaps in the gene expression regulatory networks that regulate the biological processes.…”

Application of the TGx‐28.65 transcriptomic biomarker to classify genotoxic and non‐genotoxic chemicals in human TK6 cells in the presence of rat liver S9

“…A critical gap in the application of this approach is the availability of validated gene expression signatures that can be used to predict a chemical's mode of action, or the probability that the chemical induces specific toxicities, that have been robustly tested across laboratories, cell culture models (including human models), gene expression platforms, and experimental designs. Although many studies have published transcriptional signatures to predict various toxicities [Uehara et al, ; Minowa et al, ; Cheng et al, ; Doktorova et al, ; Eichner et al, ; Thomas et al, ; Yamada et al, ; Melis et al, ; Romer et al, ; Sahini et al, ; Wei et al, ; Oshida et al, ; Schmeits et al, ; Shen et al, ], these have not been extensively validated or applied, and the majority of this work has been done on rodent cells or tissues. Therefore, accepted signatures capturing diverse toxicological targets and effects in human cells are needed for development of effective chemical screening approaches.…”

Sequence-based identification of recombination spots using pseudo nucleic acid representation and recursive feature extraction by linear kernel SVM

“…Then, a recursive feature selection approach, SVM-RFE was applied to select a group of important features for reliable identification of recombination spots. Then, through training a linear kernel SVM iteratively, the SVM-RFE algorithm is adopted to get a ranking list of all features by removing only one feature with the lowest influence on the predictions of an SVM model each time [ 26 , 27 ]. The first item in the ranking list was the most relevant feature in identification of recombination spots, and the last item had the least relevant feature.…”