Mixture Modelling for Bioinformatics Applications in Cancer Research
Kim-Anh Do
Department of Biostatistics, M.D.
Anderson Cancer Center and Graduate School of Biomedical Sciences,
University of Texas
We propose model-based inference for three different types of high-throughput
data: gene expression, mass spectrometry proteomic profiles, and phage peptide
counts. We develop Bayesian mixture models for these data, and demonstrate how
the Bayesian False Discovery Rate concept can be incorporated to handle the
problem of massive multiplicities. Applications in cancer research will be used
to highlight the performance of our methodology developments.
Kim-Anh Do is Professor in the Department of Biostatistics at MD Anderson Cancer
Center and a faculty member at the University of Texas Graduate School of
Biomedical Sciences at Houston. She is a fellow of both the Royal Statistical
Society and the American Statistical Association and a UQ Mathematics alumna.