Berberine pathway edit window. All relationships among the molecules were retrieved from the database, with this information being derived from PubMed abstracts by natural language processing technology. The function was done by selecting the data of maximum reliability by choosing all modes of interactions including romoter Binding? egulation? rotein Modification?and xpression?and by taking the relationships supported by three or more consistent data sources. Next, we picked out the incorporated genes from the imported gene list used at the onset of the pathway analysis, except the subunits of the target gene. Thus, a list of the genes associated with drug response was established with respect to not only gene expression profile data but also the biological functions of altered/ associated genes. Data from the listed approved drug library genes were used to build a support vector machine model with ArrayAssist software to predict the drug response. The SVM algorithm model with Gaussian kernels was used to distinguish sensitive cells from resistant cells, using biomarkers identified by the gene expression enzastaurin drug sensitivity correlation and pathway analysis.
The classification ability of the genes was evaluated using leave one out cross validation. RESULTS Effect of enzastaurin on the growth of lung cancer research chemicals library cells Growth inhibitory effects of enzastaurin on lung cancer cell lines were assessed by MTS assay. Figure 1 shows the sensitivity to enzastaurin among the 22 lung cancer cells. Based on the IC50, the 22 cell lines were classified into two groups, namely: enzastaurin sensitive and enzastaurinresistant. Five cell lines were sensitive, and the remaining 17 cell lines were resistant to enzastaurin. The five cell lines sensitive to enzastaurin consisted of four AC and one SCC cell line, no SCLC cell lines were enzastaurin sensitive. These results suggest that enzastaurin has anti tumour activity against NSCLC. Gene expression drug sensitivity correlation We have previously performed gene expression profile analysis of the same set of 22 lung cell lines by Affymetrix GeneChip. First, we used the MTS results for enzastaurin for the development of a molecular model of sensitivity to enzastaurin. Twenty three genes were significantly seliciclib correlated with sensitivity to enzastaurin. Next, pathway analysis was performed using the 23 genes to provide a viewpoint of the biological function of the genes, as previously described. Pathway analysis removed the incorporated genes out of the imported 23 genes.
Sixteen genes, associated with sensitivity to enzastaurin, were identified based on the biological functions of altered/associated genes. Pathway analysis national revealed that JAK1 was the final target gene for the sensitivity to enzastaurin in lung cancer cells. We next identified the optimal number of genes whose expression could accurately distinguish the sensitive cells from the resistant ones. Analysis of variance was done to remove the genes with variance. The top eight genes according to the ANOVA were subsequently found to be the minimum number necessary for prediction of drug response. We used the eight most strongly correlated genes to build an SVM algorithm model by which the five sensitive cells were distinguished from the 17 resistant cells.