Contact Orders

Plaxco K. W Simons K. T. and Baker D. Contact order, transition state placement and the refolding rates of single domain proteins. J. Mol. Biol. (1998) 277(4) 985-994. 

There does, however, appear to be a statistically significant correlation between the rate constants for folding of single domain proteins and the average sequence separation between contacting residues in the native state. Proteins that have primarily local contacts (i.e., have a low contact order) tend to fold more rapidly than those that have more non-local interactions (i.e., have a high contact order).81,82... 

Continuous-flow method 133,134, 541 Contact order 602 Control analysis 308 Control enzymes 290... 

For example, contact order (7) and long-range order (8) are... 

Fersht A (2000) Transition-state structure as a unifying basis in protein-folding mechanisms Contact order, chain topology, stability, and the extended nucleus mechanism. Proc Natl Acad Sci USA 97 1525-1929... 

On the other hand, Mirny and Shakhnovich argue that contact order has not much to do with tertiary structure, and is actually dominated by local... 

V. Villegas, J. C. Martinez, F. X. Aviles et al. Structure of the transition state in the folding process of human procarboxypeptidase A2 activation domain. Journal of Molecular Biology, 283 (1998), 1027 F. Chiti, N. Taddei, P. M. White et al. Mutational analysis of acylphosphatase suggests the importance of topology and contact order in protein folding. Nature Structural Biology, 6 (1999), 1005. 

The first column of statistics given in Table I contains the Pearson linear correlation coefficients between the descriptor values (x) and log fc/(r, iogt/). This is the statistical measure used by Plaxco et al. in their analysis of a subset of the descriptors considered here [12,14]. Consistent with their results, the two coefficients with the largest magnitudes are associated with the contact order (c and c/ ). Several descriptors not examined by Plaxco et al. [12,14] exhibit rv,iog k/ > 0.5 as well the a-helix content and propensity h andph), total helix content (a), and p-sheet content (e). Additional linear statistics are provided in Table V. Physical interpretations of the results are given in Section IV.E. 

Two of the other models in Table VI combine the contact order with a measure of the a-hehcal propensity c/n with either or p. These pairings essentially reflect the results of the previous section. The remaining model couples c/n with tic, which reveals a secondary dependence on protein size. Consistent with the sign of r , iog y (Table I), the functional dependences of log kf on these descriptors for the models in Table VI indicate that shorter proteins fold faster than longer ones (Fig. 3b). 

Figure 3. Functional dependence of calculated folding rate kf, in s ) on the normalized contact order c/n) and either (a) the normalized stability (AG/n in kcal/mol) or (b) the total number of atomic contacts tic).

The fact that the folding (and unfolding) kinetics of relatively small, two-state proteins can be predicted with reasonable accuracy from global features of the native state like the contact order, stability, and number of contacts supports the idea that the details of protein structure are not required to capture the key features of protein folding, so that reduced representations should be adequate. However, the most widely used simple heteropolymer models, those restricted to a simple cubic lattice, predict that stability is more important than native structure, in contrast to the experimental data for proteins. In this section we seek to understand why lattice models differ from proteins in this regard. Doing so is of importance because complete details of the folding trajectories of such models... 

In the present study a nonlinear, multiple-descriptor method was applied to the prediction of the logarithm of the folding rate constant for a set of 33 two- and weakly three-state proteins. With two (three) descriptors, the Pearson linear correlation coefficient between the observed values and the training set and cross-validated predictions reach 0.89 (0.93) and 0.81 (0.86), respectively. These results are to be compared with those obtained by using the contact order by itself rtm = 0.83 and r = 0.74. In addition to the contact order, some measures... 

A. R. Dinner and M. Karplus, The roles of stability and contact order in determining protein folding rates. Nature Struct. Biol. 8, 21-22 (2001). 

---