Collective motions in proteins
We develop and apply computational methods for the characterisation and comparison of the collective motions of protein structures mostly through the use of elastic network models and normal mode analysis. We implemented and maintain the WEBnm@ server which offers many useful analyses of normal modes, and also allows for comparative analyses of aligned proteins through a user-friendly interface.
Slow dynamics happen on time-scales of the order of microseconds to seconds and they are typically large-amplitude collective displacements, for example of domains. Particularly interesting are the transitions between the few stable conformations of the protein. We can learn a lot about the conformational freedom of a protein by studying the structural constraints on slow motion. Normal mode analysis (NMA) is particularly well suited to describe the conformational changes that are more easily accomodated by a certain protein structure, and we regularly employ NMA with coarse grained protein models to understand their slow motions.
Due to the functional importance of conformational changes, we are interested in learning how the slow motions involved in conformational change evolve with respect to other parameters of protein structure, and how they can help relate protein structure to function. We have used NMA to characterise domain motion in large molecular machines, to study the variation in conformational freedom within and between protein folds, and to elucidate mechanisms for protein oligomerisation. In particular we investigated the role that collective motions play on the function of a selection of proteins: calcium transport by a P-type pump (SERCA 1 Ca-ATPase), oligomeric state of evolutionary related pyrimidine operon attenuators (PyrR), enzymatic activity of acetyl-transferases (NATs) or a bacterial phospholipase C (Bt PI-PLC), and the membrane binding propensity of the latter.
To address such questions, we have been developing ways to compare the normal modes of different protein structures, and quantify their similarity. As rigorous quantitative comparison is also an important part of model validation, we have also contributed studies comparing different models for NMA.
We are maintaing a webserver for online NMA (WEBnm@), which offers many useful analyses of normal modes, and also allows for comparative analyses of aligned proteins.
Collaborators: Julian Echave, Anne Gershenson, Mary Roberts, Konrad Hinsen, Willie Taylor, Sarah Teichmann
Techniques: Elastic network models, normal mode analysis, comparative dynamics analysis