Facilities

Statistics Facilities and Qualifications

The Statistics Consulting Laboratory (Stat Lab) provides collaborative statistical expertise to University of Arizona researchers. Functions provided by the Stat Lab includes support in measurement systems (assay) development, study design, conduct, data quality assurance, graphical and numerical data analysis, and interpretation and reporting of results.

Biostatistics and Bioinformatics Facilities and Resources

The Stat Lab provides statistical services in the following areas:

Experimental Design

The Stat Lab staff work with investigators to define study objectives and endpoints, analyze pilot data, select appropriate designs, devise blinding and randomization schemes, plan interim analyses, determine early stopping rules, compute sample size with concern for the clinical as well as statistical significance, estimate the time required to accrue the total patient population, specify the statistical methods and prepare statistical sections of the protocol and grant applications. A strength of long-term Stat Lab collaboration is the involvement of biostatisticians from the initial stages of project development to the completion of the project. The Core staff function as collaborative members of research teams, which assures that every project benefits from careful scrutiny concerning appropriate study design and methods of analysis.

Conduct

Stat Lab staff work with investigators to ensure that high quality clinical and laboratory data are collected in a timely fashion, to develop data quality assurance software to check and clean data, to address unanticipated problems that arise during the study, and to monitor the study for safety.

Analysis

Stat Lab staff work with investigators to evaluate data quality, generate summary tables and figures, plan and execute interim and final analyses, and write statistical sections and results for publications. In general, Stat Lab staff adapts existing methods of statistical analysis to address scientific questions. Further, we have the capacity to develop new methods when required by the needs of specific projects.

Database Quality Assurance Service

Stat Lab staff work with investigators, Genomics Core staff, and other personnel to discuss the databases design. In addition, the Stat Lab staff will develop enhancements and refinements to existing feature sets, troubleshoot existing capabilities to ensure correctness of forms, data entry, queries and reports. New forms, queries, and reports in the database are created as needed.

Statistics Consulting Lab personnel work closely with the BIO5 Genomics Core to ensure an efficient, secure, and reliable infrastructure for data management and analysis. This coordinated approach streamlines the integration of high-dimensional sequencing data. Further, the Stat Lab’s routine use of ‚ “reproducible research” tools provides important documentation of the computational path from data acquisition through analysis results.

Personnel

The Biostatistics Core of the SWEHSC is led by Dean Billheimer, Ph.D. Dr. Billheimer is Associate Professor of Biometry and Director of the Statistics Consulting Laboratory (Stat Lab). He completed his doctoral training from the Department of Statistics at the University of Washington in 1995. He has previously held positions as Assistant Professor at the Vanderbilt-Ingram Cancer Center, and Director of the Biostatistics Shared Resource at the Huntsman Cancer Institute at the University of Utah. Dr. Billheimer has wide-ranging collaborative experience in the biomedical sciences, environmental science, and the aerospace industry. These collaborations involve statistical design of clinical trials and controlled experiments, analysis of complex data sets, and biomarker discovery and development. They also include the preparation of grant proposals and scientific manuscripts. He has published more than 40 peer-reviewed papers in a variety of journals, and has served on multiple NIH P01 and special emphasis review committees. Dr. Billheimer has extensive experience in the development and application of novel analysis methods to high-dimensional parameter problems, especially in mass spectrometry-based proteomics. His current research interests include the statistical evaluation of measurement systems, theory of normalization, and analysis methods for compositional data.

Space and Equipment

Stat Lab offices and computing facilities are located in the BIO5 Institute’s Thomas W. Keating Bioresearch Building at the UA (~500 sq. ft. The Stat Lab maintains a modern desktop computing environment via networked Unix-based and Apple computers. Data integrity is maintained by hourly incremental back-up. Internet access is provided by campus-wide 100GB Internet2. Software available on Stat Lab computers includes the R statistical software environment, python and Perl scripting languages, TeX document preparation system, as well as other Unix-based computing tools and programming languages (e.g., C++, MySQL). In addition, there is a full line of Microsoft products including Access, Excel, Word, and PowerPoint. Other statistical software available includes SAS, Stata, and JAGS, among others. The Stat Lab makes extensive use of reproducible research computing tools (described below), whereby all data management and statistical analysis code is embedded into a LaTeX document. Thus, analysis results can be computed dynamically from raw data, and automatically included in an executable document. Moreover, the data and document together constitute a history of the analysis project. In addition, Stat Lab statisticians make use of the Rshiny environment for rapid development and deployment of novel analysis tools.

In addition to a modern desktop computing environment, computing facilities include two distributed full service data centers and the data analytics and processing laboratory. The Stat Lab has dedicated access to the high performance computing (HPC) resources on the University of Arizona campus including a large shared memory (1 Tb) and multi-core clusters, and a multi-institutional distributed computing grid (Condor). For tasks requiring additional HPC resources, the SCL may utilize capacity offered by the national centers and the Open Science Grid (OSG) with seamless transition of jobs between local, regional and national resources.

Stat Lab personnel also have access to remote collaboration tools to enable virtual organizations (VO) and remote team science activities, via the Arizona Research Lab's Biotechnology Computing

Facility. These collaboration tools allow multiple research teams to conduct web-based meetings, seminars, share applications and data analysis tools from their desktop. These tools also allow audio/video-based interactions without the need of any specialized hardware on either end.