Michelle Kuttel

 

Research Projects in Computational Science

 

Computational science is interdisciplinary, requiring the collaboration of both computer scientists and domain scientists. I have a range of research collaborations in specific areas, as follows.

Current Research Projects

Anti-Bacterial Vaccines

For many bacterial species, the external capsular polysaccharides which surround the cells are essential for bacterial virulence and vaccine-induced serum antibodies against capsular polysaccharides can confer resistance to pneumococcal disease. Prof. Neil Ravenscroft (UCT Chemistry) and I have a joint research project in investigation of bacterial polysaccharides important for the development of modern vaccines, combining NMR experiments with molecular modelling methods to predict polysaccharide conformations.

Biofilms Microorganisms often organise in biofilms, where a matrix immobilises cells in close proximity to each other. Biofilms are implicated in persistent infections, particularly nosocomial infections that are difficult to eradicate. I collaborate with carbohydrate experimentalists Prof. Roberto Rizzo and Prof. Paola Cescutti of University of Trieste, Italy to investigate the conformation, dynamics and interactions of carbohydrate components of biofilms through simulation.

CarbBuilder

CarbBuilder

Characterisation of carbohydrate molecular conformation remains a central problem in glycobiology and molecular models are increasingly used to interpret experimental results, or in the absence thereof. I have a long-standing collaboration withProf. Goran Widmalm (U. Stockholm, Sweden) to develop and test our CarbBuilder research software for the building of carbohydrate models from primary structure information using effective heuristics [1,30]. CarbBuilder can generate a wide variety of carbohydrate structures, ranging from monosaccharides to large, branched polysaccharides. Future developments will focus on increasing the range of monosaccharides building blocks supported, as well as extending to automated building of glycoproteins and glycolipids.

AstroComp

I have a collaboration with Dr. Sarah Blyth (UCT Astronomy) to develop high performance computing software solutions for the new international SKA radio telescope. With Dr Anja Schroeder (SAAO) we have a specific focus on detection and mitigation of radio frequency interference.

SciVis

I have a strong interest in the principles and practice of visualisation, particularly the use of scientific visualisation to facilitate research. This interest cuts across all my projects listed above: scientific visualisation is increasingly important for exploring large data sets to test hypothesis and propose new ones. I have collaborated with research software developer John Stone (UIUC, USA) to develop novel visualisations of molecules and with Dr Sarah Blyth to develop visualisations of data cubes. In addition, I have taught a Honours-level visualization course for six years, which in 2017 will develop into a Masters-level course for the M.Sc. in Data Science.
HPC High Performance Computing (HPC) is another area of direct relevance to Computational Science: without the use of parallel computers and parallel software, much of the research I perform would be impracticable. However, I also have a direct interest in developing parallel software. I have a long-standing collaboration with Dr Robert Best (NIH, Washington, USA) to develop efficient high performance software for simulation of protein-protein interactions on Graphics Processing Units.