We prefer to ignore the usual motifs of protein secondary and higher structure, and search for an alternative, robust measure of structure. One ("contact map") approach that we have developed involved measuring effective networks of graphs whose nodes are alpha carbons in the protein chain, and whose edges link nodes within a certain effective distance. This approach, for certain distances, allows sensitive detection of the interior and the surface of a protein. Can we use it to classify different proteins?
A computational project is availabe that will explore the characterisitc length scales of our srtucutral probe for a suite of (hopefully many) proteins, analysing structural data from the Protein Data Bank (https://www.rcsb.org/) using our contact map technikque.
Interest and curiosity in application of basic concepts of structure, protein folding and graph theory is essential. Familiarity with software coding, to automate analysis from a large database is preferred.