Shankar Subramaniam: Research Interests, Protein Structure
Protein Structure: For the past several years, Subramaniam group has been working on
novel methods for protein homology modeling and structure refinement. A significant
problem in modern protein science is the apparent structural and functional similarity
between proteins despite apparent lack of sequence homology. It is apparent that there is a
hidden level of homology that conserves structure and function and Subramaniam group
has been exploring possible solutions to the hidden homology” problem. Using a
structural fragment-based approach, they have shown the limitations of standard secondary
structure approaches to homology as well as the limitations of using well-defined structural
motifs. Given that the details of atomic interactions play a dominant role in shaping the
three dimensional structure of proteins, he and his group have derived from high resolution
protein structures, pair-wise atomic distance probability density functions, which in turn can
be used to assess and refine protein structures. Using this method, low- resolution protein
structures are seen to converge to high-resolution structures. This method is being coupled
now to protein crystallography and NMR methods for rapidly obtaining high-resolution
structures.