Table 4

Associations between human-macaque neutral substitution rates and frequencies of various classes of functional elements

Class of elements

Short description

Conservation based

Reference

Correlation coefficient

Partial correlation coefficient

phyloHMM (P)

Predicted functional elements; highly conserved non-exonic sequences identified by phyloHMM

Yes (17 vertebrate species)

[52]

-0.32

-0.38

ESPERR-RP (P)

Predicted regulatory elements; non-exonic sequences with high regulatory potential, as measured by the ESPERR-RP score

Yes (7 mammalian species)

[39]

-0.24

-0.30

Enhancers (P)

Predicted enhancers; non-exonic sequences under strong constraint in human-rodent comparisons

Yes (human, mouse, rat)

[38,48]

-0.06

-0.22

CTCF-binding sites (P)

Predicted CTCF binding sites; identified by single sequence motif finding methods

No

[41]

-0.12

-0.10

CTCF-binding sites (E)

Experimentally mapped CTCF binding sites

No

[41]

-0.20

-0.08

ER binding sites (E)

Experimentally mapped estrogen receptor binding sites

No

[42]

-0.14

-0.09

RNA polymerase II binding sites (E)

Experimentally mapped RNA polymerase II binding sites

No

[42]

-0.11

0.01

Pearson's correlation and partial correlation coefficients. The substitution rates are estimated from all sites in ancestral repeats (without requiring orthology to mouse, rat, dog and cow) for each of 2,270 windows of size1 Mb.

Tyekucheva et al. Genome Biology 2008 9:R76   doi:10.1186/gb-2008-9-4-r76