Table 2

The higher the similarity between two circuits is, the higher is the overlap between their sets of alternative phenotypes

N

c

d

Mean C₁ ± S.E. (Sample size^a)

Spearman's ρ

p-value

8

0.4

0.125

0.95 ± 0.002 (9124)

-0.279

< 2.2 × 10^-16

0.25

0.956 ± 0.002 (9248)

-0.247

< 2.2 × 10^-16

0.5

0.968 ± 0.002 (9499)

-0.193

< 2.2 × 10^-16

0.3

0.125

0.933 ± 0.002 (9408)

-0.295

< 2.2 × 10^-16

0.25

0.94 ± 0.002 (9459)

-0.27

< 2.2 × 10^-16

0.5

0.963 ± 0.002 (9649)

-0.206

< 2.2 × 10^-16

20

0.3

0.1

0.968 ± 0.002 (8712)

-0.237

< 2.2 × 10^-16

0.25

0.97 ± 0.002 (8910)

-0.201

< 2.2 × 10^-16

0.5

0.98 ± 0.001 (9403)

-0.153

< 2.2 × 10^-16

0.2

0.1

0.962 ± 0.002 (9216)

-0.248

< 2.2 × 10^-16

0.25

0.965 ± 0.002 (9179)

-0.214

< 2.2 × 10^-16

0.5

0.977 ± 0.001 (9440)

-0.178

< 2.2 × 10^-16

0.1

0.1

0.948 ± 0.002 (9675)

-0.255

< 2.2 × 10^-16

0.25

0.957 ± 0.002 (9696)

-0.237

< 2.2 × 10^-16

0.5

0.966 ± 0.002 (9748)

-0.189

< 2.2 × 10^-16

This overlap decreases with the genetic distance between the circuits, according to Spearman's ρ, a non-parametric rank correlation coefficient [76]. C₁ is the overlap in the sets of alternative phenotypes of two genotypes that differ in a single regulatory interaction.

^a We sampled 10⁴ independent genotypes for each parameter combination, but we discarded genotypes that differ in k interactions where P_s0,k was empty.

Espinosa-Soto et al. BMC Evolutionary Biology 2011 11:5   doi:10.1186/1471-2148-11-5