I develop statistical methods for researchers who
seek quick and efficient analyses of new
and computationally challenging data. Consequently, I am interested
in developing methods for newly emerging data types that are motivated by
scientific discovery (e.g. SNP intensity data of related individuals).
Using both family and population based data, I have
developed improved methods for mapping genetic risk
variants for psychiatric disorders
(e.g. genes and CNVs related to Alzheimer's disease and bipolar
disorder, respectively), and for estimating genetic maps by combining
information efficiently from multiple studies (i.e. meta-analyses).
In addition, I have a general interest in deterministic
math models, gene networks, and problems that arise in
computational molecular biology.
Most of my methods employ a rich variety of deterministic and
stochastic algorithms, Monte Carlo sampling
techniques, and various optimization procedures to estimate genetic
quantities of interest (e.g. lod socres, kinships,
CNV frequencies and carrier status) in fairly general settings.
William Stewart Columbia Biostatistics Bill Stewart
Homepage
IGS : A seminar for interdisciplinary research. Last Updated: July 22, 2008
