Figure 4
![Figure 4](http://media.springernature.com/full/springer-static/image/art%3A10.1186%2F1471-2164-14-831/MediaObjects/12864_2013_Article_5538_Fig4_HTML.jpg)
Transcript clustering implies that age-related changes are of primary importance in the development of AD phenotypes. (A) Singular value decomposition of 4 and 6 months samples shows that age (column 1) is the greatest driver of variation between samples. Column 2 shows that genotype is the second greatest driver of variation. (B) This is confirmed using hierarchical clustering. (C) Comparing the 4 months samples with the 6 months samples using Cuffdiff identified 288 genes that were differentially expressed (184 upregulated, 104 down regulated, see Additional file 1: Table S1). (D) Pathway analysis using DAVID showed that members of the ribosome and oxidative phosphorylation pathways (from KEGG) are significantly overrepresented (P < 0.05) in the DE gene list.