A Quantitative Model of Transcriptional Regulation Reveals the Influence of Binding Location on Expression
Putative regulatory regions in liver and 3T3-L1 cells were identified from ChIP-seq experiments and are defined as sites bound by p300 and/or at least two other transcription factors. We performed a similar analysis in liver and cerebellum using ChIP-chip with promoter microarrays, where a regulatory region is defined as any site bound by CBP. (A) Genes with a regulatory region within 5kb of their transcription start site have a higher mean expression level than genes with no binding event. Error bars indicate +/− s.e.m. (B) Bound genes display large variation in levels of absolute gene expression. (C) Putative regulatory regions show great variation in their sequence conservation levels. Conservation level was calculated as the maximum 100bp moving average of Phastcons scores from alignments of placental mammal genomes.