Figure 2.

Explaining regulatory complexity with neutral evolution. a. Simulations using a 5-gene pathway under mutation and drift yield log-linear increases in the number of binding sites (#TFBS), degree, multiplicity, and redundancy with respect to the amount of ncDNA per transcription unit. Each line corresponds to a different binding site length, given in the legend. b. The results from running population genetic simulations on a random fitness landscape (y-axis) compared with the results from neutral - mutation and drift - simulations (x-axis). Deviations from neutrality are points off the diagonal. c. The pathway properties respond to the contributing factor s, which goes from insignificant (s=10−6) to dominant (s=1). The x-axis denotes the strength of selection in a population, or Ns. The rows denote the property under selection and the columns represent the high gain and high loss environments, respectively. The optimal value was chosen against the neutral bias and is denoted by a dashed line for degree. In the case of redundancy, the optimal value was 0 (high gain) and 1 (high loss). Shapes denote the regulatory complexity measurements of #TFBS (□), degree (○), multiplicity (×), and redundancy (△).

Ruths and Nakhleh BMC Evolutionary Biology 2012 12:159   doi:10.1186/1471-2148-12-159
Download authors' original image