Fig. 3
Schematic representation of temporal changes in differentially expressed genes and validation of microarray data of select genes by qPCR analysis. a Bar graphs representing the number of differentially expressed genes identified after microarray analysis of CL tissues collected from rats treated with VEH, AI & AI+E2 treatments. Data analyzed by Bioconductor analysis tool employing ≥2 and ≥1.5 fold change cut-offs. A comparison of total number of up regulated and down regulated genes between VEH vs. AI (orange bars) and vs. AI+E2 (green bars). The commonly regulated genes post E2 deprivation whose expression reversed with E2 replacement is represented by filled bars. b The fold change expression (microarray data) of selected up and down regulated genes associated with specific biological process at AI and AI+E2. These genes were also subjected to qPCR analysis and log2 (fold change) expression. Individual bars for each gene represents mean±SEM values in mRNA expression for microarray analysis (blue bars) and qPCR analysis (yellow bars) during different treatments (n = 3 for microarray analysis and n = 4 for qPCR analysis. Upper panel, VEH vs. AI and lower panel, VEH vs. AI+E2. *** P <0.001, ** P <0.01, * P <0.05). Individual bars for each gene represents mean±SEM values in mRNA expression for microarray analysis (blue bars) and qPCR analysis (yellow bars) during different treatments (n = 3 for microarray analysis and n = 4 for qPCR analysis. *** P <0.001, ** P <0.01, * P <0.05). c Correlation analysis between expression ratios obtained from microarray and qPCR analyses of few genes post different treatments (n = 8 genes/treatment). Linear regression analysis was performed for selected differentially expressed genes using qPCR fold change in expression values (2-ΔΔCT; Y-axis) with fold change expression values obtained by microarray analysis (X-axis). The r value, correlation coefficient, generated for the theoretical line of best fit (represented as solid line in each panel) and the P value indicates significance of the correlation as determined by ‘F’ test