Scientists join forces to tackle bovine TB

The University of Southampton's Schools of Management and Mathematics are teaming up with colleagues in the School of Biological Sciences for pioneering research into tuberculosis in cattle.

Around 20,000 cows are killed in the UK every year after contracting bovine TB. Present diagnostic tests are slow and inaccurate. Any cattle that test positive are killed and restrictions imposed on the movement of herds.

The project, funded by a £210,000 grant from the Biotechnology and Biological Sciences Research Council (BBSRC) in collaboration with the Department for the Environment, Food and Rural Affairs (DEFRA), involves the analysis of blood samples from cattle known to have the disease and others guaranteed to be free from TB.

Both sets of samples are analysed using a technique called mass spectrometry, which very accurately measures the masses of different molecules in each sample. As each blood sample results in some 15,000 individual pieces of data, the challenge is to identify components that tend to be present in the infected sample but not in the other, or vice versa. This large-scale analysis of proteins and peptides is known as proteomics.

This is where the mathematicians and management scientists come in, using sophisticated statistical analyses to remove irrelevant data and narrow down the field of suitable proteins and peptides.

Professor David O'Connor, Director of the University's Centre for Proteomic Research, heads the project, which also involves Professor Lyn Thomas and Dr Julia Bennell of Management and Professor Chris Potts of Mathematics .

"Our work centres around diagnosing the disease at an early stage," said Professor O'Connor. "It involves scientific skills, of course, but we also need help from our colleagues in Maths and Management, who are experts in extracting predictive "signatures" from large amounts of data. This needs tremendous number-crunching capabilities to process all the information."

Professor Thomas added: "We research and advise companies on the methods that estimate which consumers will default on their loans or credit cards and which will repay as required. These methods analyse a large number of variables which describe the consumers' behaviour and identify which combinations are most likely to result in default. This involves the same processes as the work on analysing the components of blood samples."

The team also works with the BBSRC Institute for Animal Health and the Veterinary Laboratories Agency. Additionally, a PhD student from the University's e-Science Centre works jointly with the Schools of Management, Mathematics and Biological Sciences on related aspects of the research.