Prediction of genotoxicity based on gene expression
Vrije Universiteit Brussel, Brussels, Belgium
In this study, new prediction models for genotoxicity were developed based on a reference dataset of 38 chemicals. Human liver cells were treated with the chemicals and resulting gene expression data were obtained using qPCR. 84 genes were selected and different machine learning algorithms were used and compared with regard to their predictive accuracy. In addition, the applicability of the prediction models was investigated on a publicly available gene expression dataset generated with RNA sequencing. To facilitate data analysis, an online application was developed, combining the outcomes of two prediction models. This research demonstrates that the combination of gene expression data with supervised machine learning algorithms can contribute towards a human-relevant in-vitro genotoxicity testing strategy.
Novel prediction models for genotoxicity based on biomarker genes in human HepaRGTM cells
Added on: 12-12-2022