Fig. 6.7 Superposition of the trajectory structures of a b-heptapeptide (see Panel A, Fig. 6.1) at 360 K with RMSD (residues 2-6) from the central structure of 0.09 nm, and a maximum RMSD between any two structures of 0.16 nm [44].

Fig. 6.7 Superposition of the trajectory structures of a b-heptapeptide (see Panel A, Fig. 6.1) at 360 K with RMSD (residues 2-6) from the central structure of 0.09 nm, and a maximum RMSD between any two structures of 0.16 nm [44].

Fig. 6.8 Number of clusters (conformers) of a b-heptapeptide (see Panel A, Fig. 6.1) at 340 K and at a pressure of 1 atm (■) and 1000 atm (•) as a function of time. In the upper panel each point represents the total number of clusters (conformers) at the corresponding time point and in the lower panel the number of clusters (conformers) that make up 95% of the trajectory sampled at the corresponding time point [23].

Fig. 6.8 Number of clusters (conformers) of a b-heptapeptide (see Panel A, Fig. 6.1) at 340 K and at a pressure of 1 atm (■) and 1000 atm (•) as a function of time. In the upper panel each point represents the total number of clusters (conformers) at the corresponding time point and in the lower panel the number of clusters (conformers) that make up 95% of the trajectory sampled at the corresponding time point [23].

this case, will determine how precisely a particular cluster is defined. Because the clustering algorithm [44] tends to produce many very sparsely populated clusters after having found the most populated ones, the convergence of the (un)folding equilibrium is better characterized by monitoring not the total number of clusters, but the number of conformational clusters that make up e.g. 95% of the trajectory sampled at the corresponding time point, see Fig. 6.8. This figure shows that the conformational space of the 7-b-peptide is basically completely sampled within about 30 ns. However, to obtain sufficient statistics on (un)folding events much longer simulation times are required as is suggested by Fig. 6.2.

Figure 6.9 shows cases in which the number of sampled conformational clusters does not level off with time, but displays a linear growth with time: the poly-hydroxybutanoate solute continuously accesses new conformations, because there are no hydrogen-bond donor moieties in this chain molecule [45]. Intrasolute hydrogen bonding does not restrict the conformational space accessible to this molecule.

Figure 6.10 demonstrates that for longer chain molecules even 100 ns of sampling at 298 K is not sufficient to find the most dominant P-2.512 helical conformer [36]. Only by simulating at higher temperature, 340 K, it was found, and subsequent simulation starting from this helical structure confirmed its dominance and stability also at 298 K. The lowest panel of Fig. 6.1 illustrates that the presence of polar side chains may slow down the (un)folding process. The conver-

Fig. 6.9 Number of clusters (conformers) of two b-depsihexapeptides which differ in the side chain structure: (•) represents the b-depsihexapeptide with all alanine residues and ( Y) represents the b-depsihexapeptide with alanine, valine and leucine side chains at 298 K and 1 atm as a function of time [45].

Fig. 6.9 Number of clusters (conformers) of two b-depsihexapeptides which differ in the side chain structure: (•) represents the b-depsihexapeptide with all alanine residues and ( Y) represents the b-depsihexapeptide with alanine, valine and leucine side chains at 298 K and 1 atm as a function of time [45].

Fig. 6.10 Backbone atom-positional root-mean-square deviation of MD trajectory structures with respect to a 2.512-helical model structure (residues 2-7) derived from NMR data for a b-octapeptide in methanol at 298 K (upper panel) and 340 K (lower panel) simulated from different starting structures. Blue and red curves: an extended peptide structure; green curve: a 2.512-helical structure [36].

Fig. 6.10 Backbone atom-positional root-mean-square deviation of MD trajectory structures with respect to a 2.512-helical model structure (residues 2-7) derived from NMR data for a b-octapeptide in methanol at 298 K (upper panel) and 340 K (lower panel) simulated from different starting structures. Blue and red curves: an extended peptide structure; green curve: a 2.512-helical structure [36].

gence of folding equilibria of a-peptides in water is very much slower than that of b-peptides in methanol [46].

Was this article helpful?

## Post a comment