Sue Wilson
Sue Wilson
The heart of statistics is data. Sue's research always has been, and
will continue to be, data-motivated. In general her research interests can
be described broadly as the development of mathematical statistical models
for data analysis to accommodate unusual or special features of the data,
as well as the implementation of these models. The breadth of topics requiring
such development is enormous. Generally, her choice of specialisation has
been motivated by the importance of a full and proper understanding of
the data in relation to its generating discipline. Often however, the resulting
methodology is applicable more broadly. Much of Sue's research has arisen
directly from extensive consulting experience in the biological, social
and medical sciences. The overall significance of her research is first
its considerable practical importance, to statistical science as well as
to society, second, the novelty of the approach taken, and third its position
on the frontiers of the important interfaces between statistical science
and the discipline from which the data originated. Sue's research has always
been positioned to have an impact on future directions of the subject.
In the following, current and recent research interests are separated into
two very broad headings, and speciality areas of research are listed under
each.
Statistical Genetics and Bioinformatics:
Development of variance component models in genetics, including analysis of data from twin and family studies; Design and analysis of gene frequency perturbation experiments; Examination of linkage and multi-locus associations; Determining locus order and genetic linkage maps; Analysis of DNA sequence data; Examining the role of environmental factors in genetics; Genomic epidemiology; Design and analysis of genome scanning linkage strategies to locate gene/s which determine complex traits; Epistasis; Analysis of data from gene expression studies; Comparative genomics; Functional genomics; Microarray analysis.
Developments in Applied Statistics:
Generalised linear mixed models: their evaluation and applications;
Design and analysis of longitudinal studies; Medical trials, including
case-control studies; Designs for cohort studies, and analysis of resultant
data; Statistical software evaluation; Graphical techniques; Use of acoustic
methods in exploratory data analysis; Sample size determination; Survival
data analysis; Analysis of physical anthropometric data; Statistical hypothesis
testing with complex survey data; Statistical modelling of the HIV/AIDS
Epidemic; Development of multistage models for the natural history of HIV
infection, allowing for changes in treatment regimes; Statistical strategy
and sensitivity analyses.