Theoretical Insights into the Biophysics of Protein Bi-stability and Evolutionary Switches
Fig 3
Rationalization of the GA/GB switch and model prediction of incremental stabilization of the alternate fold.
(a) Free energy landscapes as a function of the progress variables QA and QB are simulated in our hybrid model (εB = −0.37). The QA/QB scale (bottom-left axes for GBwt) is identical for all GA/GB variants. Free energy, in units of kBT, is the negative natural logarithm of the sampled population (Methods). For each sequence, this quantity is computed for points on a ~100×100 grid at the sequence’s melting temperature Tm. The free energies for the grid points are plotted according to the color code on the right, with the lowest free energy on the grid normalized to zero for each sequence. Note that all resulting free energy values ≥ 6 are shown in the same color. (b) Free energy differences ΔF(GA-GB). (c) Comparing sequence-dependent Tms from experiment and simulation, each normalized to the range defined by GA77 (set to 1) and GB98 (set to 0). The Tm values in (c) are in an arbitrary unit for a non-absolute temperature scale. (d) Scatter plot between absolute melting temperatures in simulation (model unit) and in experiment (in K). The experimental Tms used in the comparison in (c) and (d) are from refs. [19,20].