Abstract
Identifying drug target genes in gene expression profiles is not straightforward. Because a drug targets proteins and not mRNAs, the mRNA expression of drug target genes is not always altered. In addition, the interaction between a drug and protein can be context dependent; this means that simple drug incubation experiments on cell lines do not always reflect the real situation during active disease. In this paper, I applied tensor-decomposition-based unsupervised feature extraction to the integrated analysis using a mathematical product of gene expression in various diseases and gene expression in the DrugMatrix dataset, where comprehensive data on gene expression during various drug treatments of rats are reported. I found that this strategy, in a fully unsupervised manner, enables researchers to identify a combined set of genes and compounds that significantly overlap with gene and drug interactions identified in the past. As an example illustrating the usefulness of this strategy in drug discovery experiments, I considered cirrhosis, for which no effective drugs have ever been proposed. The present strategy identified two promising therapeutic-target genes, CYPOR and HNFA4; for their protein products, bezafibrate was identified as a promising candidate drug, supported by in silico docking analysis.
Highlights
In silico drug discovery is an important task because experimental identification and verification of therapeutic compounds are time-consuming and expensive processes
Gene expression alteration caused by treatment with a compound may be context dependent; in other words, in a cell line, the gene expression change caused by incubation with a compound may differ from that in diseases
Intersections among them are evaluated as gene sets carrying greater confidence. This strategy has been applied to the identification of genes associated with aberrant promoter methylation common among three autoimmune diseases[14], identification of genes commonly affected by two histone deacetylase inhibitors[11], and identification of reliable biomarkers considering microRNA, mRNA expression, and compounds together[20]
Summary
In silico drug discovery is an important task because experimental identification and verification of therapeutic compounds are time-consuming and expensive processes. This strategy, i.e. integrated analysis by applying PCA-based unsupervised FE to an individual dataset separately has some drawbacks: when there are no common gene sets, there is no way to proceed with the analysis there are rarely no common gene sets To compensate for this weakness without introducing weights, in the newly developed TD-based strategy here, tensors are generated using a mathematical product of a gene expression profile of drug-treated cell lines and of a gene expression profile of a disease. The inferred gene–compound interactions were found to significantly overlap with known gene–compound interactions
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
