Beyond overall protein folding characteristics, the dynamics of individual proteins is also an area of interest. For example, SNPs, single nucleotide polymorphisms, can result in single residue protein mutations and are known to play a role in both disease and drug response. Interestingly, those SNPs that play a significant role in human disease (TP53 SNPs are associated with approximately 50% of cancers), typically on only destabilize the protein structure by (1-3 kcal mol−1) [1, 2].
Analysis of our data corroborates this and helps to shed some light on the internal dynamics: 40% of SNP mutations cause a Cα-RMSD difference of between only -0.5 Å and +0.5 Å and 40% actually show a decrease in average Cα-RMSD. Binning by secondary structure, we see that turns and bends show the least amount of variation while unstructured areas show the greatest (Figure 1).
It was hypothesized that a correlation might occur between changes in Cα-RMSD and changes in residue side chain volumes due to mutations. However, no such correlation is observed for either buried or non-buried residues, regardless of the direction of the volume change (Figure 2); the mutant proteins tend to form alternative contacts that stabilize their backbones.
Further analysis of this SNP contact behavior reveals that, relative to the wild type, more than half of the SNP mutations lose between 0 and 20 total contacts while 24 mutations increase the total number of contacts (Figure 3).
To better investigate these complex effects, we are developing more data-dense analysis techniques such as detailed residue contact analysis and wavelet analysis that take advantage of the high-resolution nature of our data. Differential analysis with these and other tools is helping us identify the molecular networks by which these subtle structural changes can propagate.