Proteomics

Proteomics Group

A major focus of our research is on the application of Proteomic Fingerprinting  technologies for the development of novel diagnostics in infectious diseases. Our principal areas of interest are infections with problematic existing diagnostics, including tuberculosis (Lancet 2006 ), Human African Trypanosomiasis (Lancet 2004 , Lancet 2004 , Trends in Parasitology 2005 ), invasive fungal infections and Chagas’ disease. The underlying concept is that infectious diseases are associated with circulating patterns of proteins that more or less uniquely define that disease state.

We use SELDI-ToF mass spectrometry to obtain proteomic profiles from  

serum, other body-fluids and cellular lysates and then apply machine-learning based pattern recognition algorithms to train classifiers to distinguish cases from controls. Through identification of the informative peaks in our proteomic ‘signature’ we hope to move from mass spectra to simple bedside tests based around more conventional dipstick-based formats. For protein/peptide identification, we use mass-spectrometry based technologies, including peptide mass fingerprinting, 2D gel electrophoresis and Liquid chromatography-electrospray ionisation-tandem mass spectrometry. This translational element will form an ever increasing part of our research program.  Together with collaborators at the National Institute of Medical Research (NIMR) we are also exploring the refinement of bioinformatics tools for feature selection and pattern recognition, with a particular focus on Support Vector Machines.

A major conceptual appeal of this approach in infectious diseases is that signatures reflecting host-pathogen interactions may allow accurate discrimination between infection and disease in clinical contexts where latent disease and colonisation currently muddy the waters.