UW Madison UW School of Medicine and Public Health

CBB logoComputational Biology and Biostatistics Summer Research Program (CBB)

Faculty Mentor Candidates


Rick Chappell, Professor, Department of Biostatistics & Medical Informatics and Department of Statistics, University of Wisonsin-Madison

I work in several areas of statistics as applied to medical research. I develop and fit models of how radiation damages tumors and normal tissue for patients with cancer. Since tumor cells respond differently to radiation than other cells, this difference can be exploited to maximize the damage to one while minimizing harm to the other. I also am interested in the design of studies that determine the proper dose of drugs or radiation to give to patients with cancer. In addition, I am involved in research into various other aspects of clinical trials.

Thomas Cook, Senior Scientist, Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison

My research focuses on most statistical aspects of randomized controlled clinical trials. Specific areas include study design, especially sample size and mid-course corrections, data reporting and presentation, and interim monitoring for safety and efficacy. Delays in reporting or classification of study events such as recurrent heart attacks or cardiac procedures can affect the analysis, and I have studied methods of accounting for such delays.

Mark Craven, Associate Professor, Department of Biostatistics & Medical Informatics and Department of Computer Science, University of Wisconsin-Madison

My research interests center around machine learning and bioinformatics. In particular my current work is focused in two areas: gene regulation and information extraction. I am developing and applying machine learning methods for uncovering the regulatory mechanisms of cells. I am also interested in developing automated methods that enable text sources to be better exploited for discovery and decision making in biomedical domains. Our approach is to use machine learning methods to induce information-extraction routines from training examples.

David DeMets, Professor, Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison

The theme of my research is in the design, monitoring and analysis of clinical trials, especially the randomized controlled clinical trial. One particular interest is developing statistical methodology for data monitoring and interim analysis. In addition to statistical methodology, issues in the design of clinical trials, such as the role of surrogate outcome measures and multiple outcomes are being considered.

Marian Fisher, Distinguished Scientist, Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison

My interests are methods to support data monitoring committees, independent groups that monitor accumulating data for safety and efficacy for ongoing large multi-center clinical trials designed to provide confirmatory evidence to gain FDA approval of a drug. I am particularly interested in graphical techniques for presenting results that allow clinicians to integrate large amounts of comparative information quickly. I have worked on clinical trials in diverse areas such as ophthalmology, cardiovascular disease, trauma, and Lou Gehrig's disease.

Ronald Gangnon, Associate Professor, Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison

My primary research interest is spatial modeling of disease rates, especially methods for proactively detecting "hot spot" clusters and for reactively assessing the significance of reported clusters. I am currently working to develop models that assess our uncertainty about the location and size of a cluster. Other research interests include order-restricted non-parametric methods for assessing the repeatability of measurements and methods for group sequential monitoring of multiple endpoints in clinical trials.

Sunduz Keles, Associate Professor, Department of Statistics and Department of Biostatistics & Medical Informatics, University of Wisconsin-Madison

Our research group is broadly interested in developing and applying statistical and computational methods for problems arising in genome biology and computational biology. Currents specific topics include statistical methods for high throughput sequencing data, regulatory motif finding in biological sequences by utilizing multiple sources of experimental and genomic data, statistical methods for analyzing data from tiling arrays (ChIP-chip, tiling expression, nucleosome occupancy, histone modification), and comparative genomics.

Christina Kendziorski, Associate Professor, Department of Biostatistics & Medical Informatics Affiliate Faculty, Department of Statistics, University of Wisconsin-Madison

My research is focused on methods to identify genes involved in disease pathogenesis. I have been particularly interested in microarray studies, which measure gene expression for thousands of genes - often an entire genome - simultaneously. With this type of data, the number of measurements of distinct genes across an array greatly exceeds that for any individual gene (large p, small n). This poses a number of interesting statistical problems in experimental design and analysis.

KyungMann Kim, Professor, Departments of Biostatistics & Medical Informatics and Statistics, University of Wisconsin-Madison

My research focuses on group sequential methods for data and safety monitoring and early stopping of clinical trials in chronic diseases such as cancer, cardiovascular diseases and human immunodeficiency virus (HIV) diseases and on regression methods for analysis of paired ordered categorical data such as from bilateral eye diseases.

Bret Larget, Associate Professor of Statistics and of Botany

My research focuses on the development of statistical methods to analyze genetic data in order to estimate phylogenetic trees while accounting for uncertainty. I am particularly interested in developing models for the evolution of DNA that incorporate current understanding of important biological processes but are feasible for statistical analysis. The Bayesian approach to statistical phylogenetics is particularly appealing, because probability is a natural and easily understood way to express uncertainty in estimates and strength of belief in various hypotheses. The computational approach of Markov chain Monte Carlo allows for practical Bayesian analysis with complicated and realistic models.

Mary Lindstrom, Professor, Department of Biostatistics and Medical Informatics, University of Wisconsin-Madison

My research focuses on methods for the analysis of sets of curves. This type of data occurs frequently in the biological sciences. For example, we might obtain a dose response curve from each of a number of people (or mice, or cell lines) which fall into two groups (say treatment and control). I work on methods that allow us to compare the groups even if there is no parametric model available (e.g. a polynomial) which fits the curves.

John Markeley, Steenbock Professor of Biomolecular Structure, and Director of the National Magnetic Resonance Facility at Madison and the BioMagResBank.

The central theme of our research is the application of nuclear magnetic resonance (NMR) spectroscopy to the solution of biochemical problems. The unique power of NMR lies in its ability to provide detailed chemical and structural information at an atomic level about molecules in solution--even when they are present in living cells or organisms. The general strategy is to use multidimensional (2D, 3D, and 4D), multinuclear magnetic resonance techniques to detect and assign resonances from atoms of biological interest (e.g., 1H, 13C, 15N, and 31P). With these assignments in hand, we can then interpret the wealth of spectral information present in coupling constants, relaxation rates, cross-relaxation rates, and chemical shifts.

Michael Newton, Professor, Department of Biostatistics & Medical Informatics and Department of Statistics, University of Wisconsin-Madison

My research considers statistical problems in the biological sciences, especially problems that involve the analysis and interpretation of genetic data. Questions range from how to use molecular sequence data to reconstruct patterns of evolution, to how to characterize genes affecting cancer risk, or to how to infer trends in wild animal abundance. Central to my research is the development and the theoretical analysis of probability models and computational statistical methods.

Rick Nordheim, Professor, Department of Statistics, University of Wisconsin-Madison

An important ecological concept is species diversity. Examples to which this concept can be applied include birds in a forest or bacteria in the stomach. Certain distributions are used to describe the pattern of population size of species and the number of species. Of particular interest are certain indices that are used to summarize diversity. The most common one of these is probably the Shannon Index which is related to entropy ideas. Most ecological studies provide estimates of such diversity indices without consideration of the uncertainty in their estimation and how sampling strategies -- and the underlying species distribution -- affect the estimates. We will explore various ways of quantifying the performance of these diversity indices.

David Page, Professor, Departments of Biostatistics & Medical Informatics and Computer Science, University of Wisconsin-Madison

My research is focused on machine learning and data mining with applications to bioinformatics, chemoinformatics, and health sciences. Of particular interest, are techniques such as inductive logic programming that can utilize background knowledge and return human-comprehensible results, and other relational learning techniques capable of dealing with complex data points (such as molecules) and producing logical rules.

Nicole Perna, Assistant Professor, Animal Health & Biomedical Sciences, School of Veterinary Medicine, University of Wisconsin-Madison

Our research is directed at understanding the molecular evolution of complex traits like virulence and host range. Our work focuses on the Enterobacteriaceae, a bacterial family with members occupying diverse environmental niches including plant and animal hosts with lifestyles ranging from mutualism to pathogenesis. We use both computational and experimental approaches to study the rates, patterns, mechanisms and phenotypic consequences of genome-scale evolution. Some of our current areas of research include 1) Specialization in either plant or animal hosts, 2) Recombination within and between species, populations and strains, 3) Rates and patterns of genome rearrangement, 4) Regulatory evolution and energy metabolism, and 5) Evolution of chemotaxis and signaling systems.

Marjorie Rosenberg, Professor, Department of Actuarial Science, Risk Management and Insurance Department, School of Business, Department of Biostatistics & Medical Informatics, Medical School, University of Wisconsin-Madison

My research interests are in the application of statistical methods to health care, and applying my actuarial expertise to cost and policy issues in health care. I am involved in the development of a cost-effectiveness model to determine whether newborns should be tested for cystic fibrosis. I am also on the faculty and member of the evaluation team for the University of Wisconsin Center of Excellence in Women's Health. Finally, I am interested in the use of insurance claim data in the development of statistical tools that can monitor health care outcome processes.

Jude Shavlik, Professor Departments of Computer Sciences and Biostatistics & Medical Informatics, University of Wisconsin-Madison

My primary research interests are machine learning and datamining applied to biomedical tasks such as microarray ("gene chip") analysis and design, protein-structure determination, and information extraction from on-line biomedical text. Within machine learning I am particularly interested in the use of prior knowledge (i.e., going beyond the tradition of only using labeled examples), learning "human-readable" models, applications of inductive-logic programming, and the use of ensemble methods.

Grace Wahba, Bascom Professor of Statistics, Professor of Biostatistics and Medical Informatics, Professor of Computer Sciences (by courtesy)

My research involves multivariate function estimation and machine learning using splines, support vector machines and variational methods. I'm particularly interested in developing models relating risk factors to health outcomes in large medical and environmental studies. I'm also involved in studying risk factors for eye diseases in large demographic studies, where several thousand selected volunteers are followed over a period of time.

Brian Yandell, Professor, University of Wisconsin-Madison

My research interests emerge from collaboration with biological scientists, particularly through my Biometry appointment at UW-Madison. Recent work has been largely in statistical genetics, with some attention to ecology. Ongoing statistical genetics research involves data analysis, methodology and design; Bayesian inference for QTLs using Markov chain Monte Carlo methods to estimate the joint distribution of QTL number, locations and effects; combining genetic data across multiple experiments and semi-parametric and non-parametric inference for QTLs. Other theoretical and applied genetics research concern mixed models and polyploid genetics. Recent collaborations with Gianola (Animal Science) examines genetic map construction. Considerable recent research has been focused on microarray gene expression data analysis, particularly with Lin (Statistics) and Attie (Biochemistry) on microarray data analysis. Research in ecological modeling builds on novel ideas about individual-based models in population ethology with a software release.


Any questions about the program or application procedures?

Please contact: Whitney Sweeney, Student Services Coordinator
Telephone: 608-262-9184 or Email:


Last Modified:
Copyright © 2014 Board of Regents - University of Wisconsin System