University of Cambridge > Cambridge Veterinary School > Cambridge Infectious Diseases Consortium > People

Cambridge Infectious Diseases People

Name:

TJ McKinley


Position(s):

Postdoctoral statistician


Email:

tjm44@cam.ac.uk


Tel.:

+44 (0) 1223 337685

Research description

I am interested in applying and developing statistical methodology for the study of infectious diseases. Inference for epidemic models is challenging, since we often only have partially observed data, and the construction and calculation of the likelihood can be far from trivial. This is further compounded when modelling at a large-scale, such as is being done for current models of avian influenza, foot-and-mouth disease (FMD) or bovine tuberculosis. Of particular interest in my research is the application of Approximate Bayesian Computation (ABC) techniques as an alternative means of parameterising epidemic systems.

ABC approaches utilise the fact that although the likelihood may be difficult to calculate, the underlying model may in fact be straightforward to simulate from. In this case standard Markov chain Monte Carlo or Sequential Monte Carlo techniques can be adapted to include an approximation to the likelihood ratio generated by comparing repeated simulations of the underlying model to the observed data. However potential advantages in computational efficiency and tractability must be balanced with the loss of accuracy due to the approximation process.

I am also working on a project in collaboration with VLA exploring risk factors associated with persistent bovine tuberculosis (bTB) infections in cattle herds in the UK. Previous cross-sectional case-control studies have looked at bTB breakdowns, but there is increasing evidence that some herds may experience prolonged or repeated breakdowns over time. These herds have the potential to act as new sources of infection to other herds, either through cattle movement or possibly to source infection in local wildlife. Identifying risk factors associated with these types of breakdown could help to inform better management practices and/or control policies, and may help to provide useful insight into bTB persistence in the UK.

In addition to this I also have some interest in bacterial and viral infections, and have been involved in projects exploring in vivo and in vitro dynamics of Salmonella infections, and natural transmission chains of equine influenza virus through its natural host. Furthermore I have also been involved with a project examining risk factors for parietal and visceral pleurisy in pigs in the UK.

Background

I graduated from the University of Exeter in 2003 with a BSc (Hons.) in Mathematics and stayed on to study for a PhD in Statistics under the supervision of Professor Trevor Bailey. My thesis title was, 'Spatial survival analysis of infectious animal diseases', and explored the feasibility of using spatial survival analysis as a means of predicting the path of infection in large-scale infectious animal disease epidemics. The work was motivated in particular by the 2001 UK foot-and-mouth disease (FMD) epidemic and was funded by a EPSRC/VLA CASE studentship no. 0305. I passed my viva in May 2007.

Main collaborators

James Wood, Kat Karolemeas, Andrew Conlan, Olivier Restif (all CIDC), Rob Deardon (University of Guelph, Canada), Alex Cook (Dept. Plant Sciences, University of Cambridge), Josh Ross (University of Adelaide), Andy Mitchell, Tony Goodchild (VLA), Andrew Grant, Duncan Maskell, Pietro Mastroeni, Clare Bryant (all Dept. Vet. Med., University of Cambridge), Julia Gog (DAMTP, University of Cambridge), Trevor Bailey (SECaM, University of Exeter), Mark Arnold (VLA), Peter Durr (CSIRO, Australia), Dan Tucker and Riki Jaeger (Dept. Vet. Med., University of Cambridge).

Key publications since 2001

Published:
 

© 2010 CIDC, University of Cambridge
Information provided by webmaster@thing.cam.ac.uk