Dr. Jeffrey S. Morris
The University of Texas M. D. Anderson Cancer Center
Department of Biostatistics and Applied Mathematics
- Functional data analysis
- Linear mixed models
- Nonparametric regression
- Multivariate methods
- Bayesian methods
- Colorectal cancer
The long-term objective of my research is to develop new statistical tools to help better model complex biological data, specifically in the setting of cancer research. My most recent work includes developing methodology for modeling functional data, data for which the ideal units of observation are curves. This type of data is increasingly encountered in biomedical research as automated technologies for obtaining biological measurements are developed. I have a recently funded R01 to develop methodology for such data. My most recent work provides a generalization of linear mixed models to functional data, where the functions may be irregular and require nonparametric modeling. These methods have already been applied to colon biomarker data from animal carcinogenesis studies, MALDI-MS proteomics data from cancer studies, and accelerometer data for objectively quantifying children’s activity levels in the Planet Health intervention study at Harvard University. Other recent work includes methodology for pooling data from microarray studies conducted at different institutions using different versions of the Affymetrix oligonucleotide array, methodology for analyzing SAGE (serial analysis of gene expression) data, and methodology for preprocessing and analyzing MALDI-MS (matrix-assisted laser desorption and ionization mass spectrometry) proteomics data.
A tutorial in our laboratory would provide advanced training in statistical analyses, mathematical modeling, and computer programming.
Program in Biomathematics and Biostatistics