Table 2.

Linear regression models to predict human-chimpanzee divergence and human diversity

	Human-chimp divergence					Human diversity
	R²	Slope	τ	R²	Slope	τ
CpG islands		−0.486^*	−0.142^*		−0.337^*	−0.027
Recombination rate (cM/Mb)		0.189^*	0.248^*		0.268^*	0.270^*
Simple repeat content		0.156^*	0.299^*		N.S.	0.161^*
The top three predictors	0.242			0.150
CpG content		2.007^*	−0.023		1.623^*	0.050^**
Distance to centromere		−0.123^*	0.017		−0.100^***	0.013
Distance to telomeres		−0.123^*	−0.280^*		−0.094^**	−0.229^*
Fraction of genes expressed in testes		−0.041^**	−0.121^*		N.S.	−0.065^**
GC content		−2.445^*	−0.076^***		−2.217^*	0.006
LINE count		−0.053^**	−0.136^*		N.S.	−0.111^*
Gene content		−0.133^*	−0.207^*		N.S.	−0.073^***
Poly(A/T)		−1.023^*	0.053^**		−1.044^*	−0.016
Poly(CA)		−0.100^**	0.147^*		N.S.	0.013
Poly(G/C)		−0.149^***	−0.113^*		−0.184^***	−0.031
Full model	0.452			0.258

This model was determined based upon a stepwise procedure. The divergence and diversity data come from repetitive regions. Most of the variability is explained by the first three predictors. The R² estimates are for the model with three predictors and for the full model, that is, all 13 (8) predictors. The slope estimates are for the full model and are standardized for ease of comparison. Also listed are the pairwise rank-correlation coefficients based on Kendall's τ.

(N.S.) Not significant in the multiple regression model.
↵* Significant at the <0.1% level.
↵** Significant at the 5% level.
↵*** Significant at the 1% level.

Why do human diversity levels vary at a megabase scale?