In highly stressed and crowded cellular environments, proteins fold spontaneously into specific, biologically functional "native" forms. It can be dauntingly difficult to model this process under realistic conditions that take into account pH and other chemical potentials set by many solutes. Be they stabilizing or denaturing agents, our ability to model the action of such solutes depends on efficient computation to deliver answers in a reasonable time. 

Combining computer computation with theoretical methods of solution chemistry has let us develop a new simulation method for swift modeling of protein folding under test-tube and cellular solution conditions, rather than the artificial conditions of most simulations. The idea is to compute analytically the free energy changes associated with solute on each particular protein configuration. Then, numerical computation is needed only to sample the set of configurations as efficiently as when no bathing solute is present.

Compared to traditional methods, our method typically speeds up the calculations at least 100-fold. We have illustrated this computational strategy by considering examples of model proteins. Our goal was to reveal the impact of the myriad of large and small solute molecules found in living cells on protein stability and conformation once solute and protein interact. By comparing protein conformations - both native and denatured - and the corresponding levels of associated solute in the presence of different solutes, it becomes clear how protein structure is determined by the extent and characteristics of protein-solute association. As solute associates with protein, either favorably or unfavorably, new stable protein configurations become predominant. These protein structures or "intermediates" sensitively depend on the interactions and concentrations of solutes.  

Related References

D. Harries, V.A. Parsegian, Gibbs adsorption isotherm combined with Monte Carlo sampling to see the action of cosolutes on protein folding, Proteins 57, 311-321 (2004). [PDF]