Medical College of Wisconsin Biostatistics Research

Bayesian statistical inference is a compelling alternative to the frequentist paradigm. Recent advances in computational performance has led to the rapid adoption of Bayesian parametric and nonparametric inference within biostatistics. Bayesian methods are particularly well suited to methodologic research and applications in modern biostatistical areas such as causal inference, high-dimensional data, survival analysis, and experimental design of clinical trials. Several division faculty members specialize in Bayesian biostatistics research including Anjishnu Banerjee, Yan Gao, Prakash Laud, Brent Logan and Rodney Sparapani.

Professor Brent Logan, working in the area of multiple comparisons, has interests in clinical trials with multiple end points and multiple decisions. He has proposed designs for randomized phase II clinical trials in which one is interested in evaluating several new treatments prior to a comparative phase III clinical trial. Professor Logan provides biostatistical support to a multicenter clinical trials network for blood and marrow transplant research.

Michael Martens, PhD has developed methodology for group sequential and adaptive designs for confirmatory clinical trials that can incorporate covariate information to boost efficiency. He has also proposed techniques for toxicity monitoring of multiple safety endpoints and cohorts in clinical trials.

The Medical College of Wisconsin is a leading institution in the area of functional magnetic resonance imaging (fMRI) research with interdisciplinary research involving several departments such as Neurology, Psychiatry, Radiology and Biophysics. From the Division of Biostatistics, Professor Brent Logan, Kai Yang and Mei-Jie Zhang are conducting collaborative and methodological work in this area. Professor Zhang has analyzed fMRI data using covariate adjusted ROC curves, working with Professor Lee of the Biophysics Department. Professor Logan has compared individual voxel thresholding methods for identifying active voxels in single-subject fMRI datasets. Professors Rowe and Logan have proposed a way to directly model the complex valued fMRI data, rather than just the magnitude data as is typically done in fMRI analysis.

Model selection is one of the central issues in the application of statistical methods. It includes comparison of two models, variable selection in linear and generalized linear models, and model checking via goodness of fit tests as well as diagnostic statistics and plots.

The problem of multiple comparisons arises when performing many hypothesis tests, in which case the “familywise error rate” or chance of one or more incorrect significant findings among all these tests increases with the number of tests being performed. Professor Logan has a keen interest in this area in which he continues to publish new methods and applications. Working with Professor Rowe, Professor Brent Logan has investigated multiple comparison thresholding techniques in single-subject fMRI analysis. Professor Logan's work with Professor Tamhane at Northwestern University includes the problem of comparing two treatments at multiple endpoints and multiple comparison procedures to identify the minimum effective dose and/or maximum safe dose in the dose-response setting. In collaboration with Professor Mei-Jie Zhang, Professor Logan has investigated methods for controlling the familywise error rate when performing pairwise comparisons among several groups when the outcome is the time to an event of interest.

The Division's research in statistical genetics is led by Tao Wang, PhD and Chien-Wei Lin, PhD, with recent work includes various genetic data analyses, linkage and association mapping of disease genes, modeling and methodology development for association mapping of quantitative trait loci, and haplotype association analysis of single nucleotide polymorphisms (SNPs). Specifically, Tao Wang, PhD, explored the definitions and properties of additive, dominance and epistatic effects of QTL and partition of genetic variance in an equilibrium as well as in a disequilibrium population. Additional work resulted in the development of a population-based multipoint LD method for fine mapping of quantitative trait loci. A mixture model was applied to describe the relationship between phenotype and QTL genotypes. An EM algorithm was developed to estimate the genetic effects of QTL and joint haplotype frequencies of QTL and markers. Chien-Wei Lin, Ph.D, is interested in developing methodology in statistical genetics, translational bioinformatics, microbiome, power and sample size calculation tool for NGS data, integrative/meta-analysis methods for different omics data, and supervised/unsupervised machine learning applications. For research interests in power and sample size calculation tool, he has focused on two applications in RNA-Seq and Methyl-Seq data. He has worked closely with collaborators from different fields, including cardiovascular epidemiology, psychiatry, and cancer biology. He has experiences in various types of omics data, including single nucleotide polymorphism (SNP), copy number variation (CNV), DNA methylation, gene expression, proteomics (peptide), and metabolomics data. Paul L. Auer, PhD is broadly interested in the genetics of human diseases. His work has explored the use of genotype imputation in diverse ancestries and he is currently working on methods for incorporating genotype uncertainty into association tests. He has led multiple efforts to map rare-genetic variation and complex traits, with a focus on hematologic traits and cancers. Recently, Dr. Auer has been working with whole-genome sequencing (WGS) data from the Trans Omics for Precision Medicine (TOPMed) program and is leading an effort to analyze somatic chromosomal alterations from the TOPMed WGS data. 

The Division is developing an international reputation as a locale for statistical research in the area of survival analysis and longitudinal data analysis. This area is anchored by the collaborative work of Mei-Jie Zhang, PhD with the researchers at Department of Biostatistics at the University of Copenhagen. Other faculty that have contributed to this area include: Brent Logan, PhD, Kwang Woo Ahn, PhD, Ruta Brazauskas, PhD, Rodney Sparapani, PhD, Soyoung Kim, PhD, and Noorie Hyun, PhD.

Michael Martens, PhD, has published methods for covariate-adjusted survival and competing risks analysis within clinical trials with interim adaptations. He has developed concise and accurate sample size formulas for studies employing survival and competing risks regression models with a complex correlation structure among covariates.

Variable selection is one of the central issues in the application of statistical methods. Faculty members Kwang Woo Ahn, PhD, Soyoung Kim, PhD, Chien-Wei Lin, PhD, and Kai Yang, PhD have been working on variable selection for the generalized linear model and survival/competing risks models with possibly high dimensional data.

Michael Martens, PhD, has applied Bayesian parametric and nonparametric models for generating knockoff features to permit variable selection with guaranteed false discovery rate control.

Personalized medicine has become an important research area in biostatistics. The main goal is developing a statistical tool to identify the optimal treatment given each patient’s characteristics. Faculty members Prakash Laud, PhD, Brent Logan, PhD, and Rodney Sparapani, PhD have been working on BART (Bayesian Additive Regression Trees) with applications to survival analysis and personalized medicine. Relatedly, Kwang Woo Ahn, PhD and Soyoung Kim, PhD have been working on this area using the propensity score for survival and competing risks outcomes.

The Medical College of Wisconsin Biostatistics Consulting Service provides statistical support to biomedical investigators. Faculty and students often work with the Consulting Service to provide researchers assistance with research studies. Services offered include planning, data collection, analysis, and reporting.

View the Biostatistics Consulting Service website

The Biostatistics Shared Resource (BSSR) works collaboratively with cancer research investigators in providing analyses, computational methods, models and algorithms. In this way, it helps researchers from concept development to the publication of results and design of subsequent research. The resource’s involvement substantially contributes to advancing research and spans numerous cancer center project areas, including basic, translational and community-engaged research.

View the BSSR website

The Blood and Marrow Transplant Clinical Trials Network (BMTCTN) is a multi-center network funded by the NIH and NCI to implement clinical trials in the field of hematopoietic stem cell transplantation. It was established to conduct large multi-institutional clinical trials. Trials address important issues in hematopoietic stem cell transplantation (HSCT), thereby furthering understanding of the best possible treatment approaches. Participating BMT CTN investigators collaborate through an organization designed to maintain continuity of operations, to facilitate effective communication and cooperation among participating transplant centers and with collaborators at the National Institutes of Health, and to offer trials participation to patients in all regions of the U.S. Corporation.

MCW Biostatisticians support the clinical trials network through the Data Coordinating Center (DCC) in terms of designing the clinical trials and analyzing the trial results.

Learn More

The Center for International Blood and Marrow Transplant Research (CIBMTR) was formed by a merger of the National Marrow Donor Program in Minneapolis and the International Blood and Marrow Transplant Registry (IBMTR) at the Medical College of Wisconsin. It has enjoyed a positive, collaborative association with the Division of Biostatistics in the MCW Institute for Health & Equity since 1980, an association that is a distinctive asset and crucial to the success of CIBMTR research. Biostatisticians ensure the statistical integrity of CIBMTR scientific activities, contribute to results in articles on hematopoietic cell transplantation (HCT)-related statistical issues for clinical audiences, and support Working Committee study investigators in developing scientific study protocols using CIBMTR data. CIBMTR biostatisticians have pioneered novel methodological approaches to analyzing HCT data.

HCT is a complex process with multiple competing risks and dramatic changes in the risks of specific events over time. The CIBMTR has developed and evaluated the statistical models used in HCT research and helped guide the research community in appropriate application and interpretation of these sophisticated models.

Re-introduced in July 2018, the Center for Advancing Population Science (CAPS), formerly the Center for Patient Care and Outcomes Research (PCOR), develops, tests and implements innovative strategies for transforming healthcare that optimizes quality, value and cost. Through innovative research, analysis, implementation and impact, CAPS is set to become a global leader in healthcare transformation. With a focus on population science and global health, enhanced faculty and collaborator recruitment, and a desire to improve community engagement, CAPS conducts research on patient care services and related health outcomes, facilitates a supportive environment for new MCW investigators, determines the need for and recruit new faculty in targeted methodologic areas, and sponsors a health services research seminar series for the exchange of ideas. Professors Prakash Laud and Duo Yu  from the Division of Biostatistics work with physicians and other medical researchers in CAPS. Most of the projects here are funded by government agencies such as the National Institutes of Health and the Department of Defense or by private foundations.

Professor Szabo is the Biostatistician for the Translational Research Unit of the Clinical and Translational Science Institute (CTSI). The Biostatistics CTSI Key Function includes:

Learn More

The Mellowes Center at the Medical College of Wisconsin provides academic support for researchers at MCW who use the genomic sequence to understand disease and translate this information from the laboratory to the patient. Most of the research projects in the Center are funded by government agencies such as the National Institutes of Health. The research areas include various directions in genomics, high throughput sequencing and the development and use of single nucleotide polymorphisms (SNP's), microarray analysis and bioinformatics. Professor Tao Wang is associated with this Center.

Professor Brent Logan serves as the biostatistician for the corporate activities of the National Marrow Donor Program® (NMDP)/Be The Match® National Marrow Donor Program (NMDP) in Minneapolis. The NMDP is NMDP/Be The Match is a global leader in bone marrow transplantation. They conduct research to improve transplant outcomes, provide support and resources for patients, and partner with a global network. All centers in their network must meet quality standards. These standards are put in place to make sure that donors and patients receive high quality care and government standards are met.

Projects of statistical interest include projections of the optimal registry size and composition and development of some means of grading performance of NMDP centers in terms of patient survival.

Learn More