Root-mean-square-deviation potential energy function

Figures 18-20 summarize the results from both simulations. In Figure 18 we have plots of the potential energy as a function of simulation time, root mean square (RMS) deviation as a function of simulation time and the radius of gyration as a function of simulation time. The potential energy plot shows that the structures are well equilibrated and stable over the 100 ps. The root mean square deviation plot indicates that the final structures from both simulations are very similar. From the radius of gyration plot one can see that during the implicit solvent simulation the structure is similar to the explicit solvent simulation radius of gyration, then around 50 ps the structure contracts for about 40 ps before expanding back to the explicit solvent simulation structure.

Its practical advantage is that it avoids the use of the polarizabilities a its disadvantage is that it has no theoretical justification and that it is inaccurate. Kramer and Herschbach (1970) have compared the two combination rules with a set of accurate published calculations of 153 unlike coefficients. The root mean square deviation from experiment was 3.25% when Eq. (14) was used, but 73.5% when Eq. (15) was used Whether the use of Eq. (15) is justified depends on estimates of the overall accuracy of the various contributions to the total set of potential energy functions, a subject to which we shall return in the next chapter. 

---