Table 1.

Error in relative ΔG^el between conformational states C₁ and C₂ of various realistic structures as specified below. The error, in kcal∕mol, is computed as error$=(\Delta G_{\mathrm{GB}}^{\mathrm{el}}\left(C_1\right)-\Delta G_{\mathrm{GB}}^{\mathrm{el}}\left(C_2\right))-(\Delta G_{\mathrm{PE}}^{\mathrm{el}}\left(C_1\right)-\Delta G_{\mathrm{PE}}^{\mathrm{el}}\left(C_2\right))$. A set of snapshots representing the four conformations of ala10 used in this analysis is different from the one used for finding optimal values of (γ₀, μ₀).

Conformations C1, C2	Alpha, pp2	Left, pp2	Hairpin, pp2	Alpha, hairpin	Protein-A: folded, unfolded
Error, Ψ = 1	5.0	8.1	3.3	1.2	37.8
Error, canonical GB	1.7	1.8	2.7	−0.16	2.8
Error, proposed model	−0.1	0.07	0.5	−0.17	0.7