Skip to main content
Figure 1 | BMC Bioinformatics

Figure 1

From: Human promoter genomic composition demonstrates non-random groupings that reflect general cellular function

Figure 1

Cluster analysis of gene expression data set from mitogen stimulated T-cells compared to promoter TFBS composition. (a) K-means cluster analysis of cDNA expression profiles of phorbol ester and ionomycin stimulated Jurkat T-cells collected at 0, 1, 2, 6, 12, and 24 hours after stimulation [11]. Total genes in each cluster is indicated in parentheses. (b) Centroid plot representing average kinetic profiles of the four clusters at the six measured time intervals. (c) Principal component analysis (PCA) of TFBS frequencies in the genomic sequences extracted from the 7,298 genes profiled in Figure 1a. (1200 base pairs upstream and 200 base pairs down stream from the start of transcription). Prior to analysis, each gene was color-coded by its respective cluster shown in Figure 1a (red = cluster/group 1, no change, green = cluster/group 2 early elevated expression, blue = cluster/group 3, repressed expression, and yellow = cluster/group 4 late elevated expression). (d) The extracted promoter sequences of each gene were then compared with respect to TFBS composition alone by K-means clustering. Nine out of sixteen clusters contained more the 4 genes (indicated as groups 1,4,6,9,10,12,13,15, and 16).

Back to article page