Application of statistical methods to predictive biology

Abstracts are not currently available in GtR for all funded research. This is normally because the abstract was not required at the time of proposal submission, but may be because it included sensitive information such as personal details.

The application of an appropriate statistical method is key to extracting maximum information from a data set to answer biological questions and predict system behaviour and its associated uncertainty. Biological systems always show variability, and statistical methods can be used to identify and quantify this variability and separate it from systematic trend. This project improves the use of applied statistics to answer biological questions via data from designed experiments and observational studies. The approaches taken will include the adoption and evaluation of new statistical methods, as well as the adaptation of existing techniques into new areas of science.

Areas to be investigated include: modelling of genotype x environment interaction at both the phenotypic and genotypic levels, with evaluation of field-based designs; use of analysis of distance procedures for quantification of differences in complex multivariate data (eg. PLFA); evaluation of smoothing splines and non-linear mixed models for responses from field and laboratory data; estimation of parameters within general process models; model validation, particularly stochastic simulation models; detection of data outliers within complex linear models.

unavailable