The Biostatistics Core will ensure that Center investigators will have sufficient availability of biostatistical and bioinformatics expertise to complete the work on a timely basis and allow necessary interchange with Center investigators. The Biostatistics Core investigators will meet with project investigators regulariy to discuss pertinent analyflc and data management issues. Statistical input in the eariy stages of study design for speciflc aim 1 and data management in speciflc aim 2 will increase the efficiency of investigations and improve the quality of research. These include selection of appropriate study design, sample size and power calculation, missing data analysis, identiflcation of sources of bias, dealing with confounding, and esflmaflon and interpretation of qualitative and quantitative interaction. Both exploratory analysis and advanced modeling approaches for clinic and basic core studies in speciflc aim 3 will allow us to examine the effects between exposure-disease and to test the associations among disease progression across time, pain experience, treatments, psychosocial variables, intervention, demographic variables and other confounding variables as well as baseline measures. Geneflc analysis and bioinformaflcs support for high-throughput microarray gene expression studies in speciflc aim 4 will help the investigators to identify unique geneflc signatures and biomarkers that are up, down or not regulated before and to have a better understanding of the complex pathogenesis and peripheral mechanisms of gene-gene and gene-environment interactions on the pain outcome and to develop better common intervention strategies for the pain management. Two-interconnected mixed model provides us the gene-bygene variance estimates and it accounts for inter-flme variability as well as human/mice sample variability; therefore they may improve the accuracy of the results and achieve better power. Trajectory analysis intends to identify novel dynamic molecular patterns that could cause neuronal degeneraflon and to examine the monotonic changes from time to time for the selected differentially expressed genes [Holter, et al. 2000]. Pathway analysis and funcflonal analysis allow us to further examine these differenflally expressed genes and their interactions by incorporating the expert knowledge and published results. Association analysis will correlate and integrate the results from both animal and human studies in order to gain significant increases in power of the overall study and more importantly to resolve the disconnection between human and animal studies. The classification and prediction models will not only serve as validation methods for the selected risk factors but also can be used for predicting individuals with high risk of diseases progressions and pain development. The Biostatistics Core team will include Dr. Liang and Dr. Fang. Dr. Liang is an Associate Professor of Biostatistics in the Department of Organizational Systems and Adult Health and a senior biostatistician for the research office of SON at UMB. She will provide leadership in the development of study design, statistical and genetic analysis plan, and novel methods for all projects. Dr. Liang has expertise in research at the intersection of genetic marker and gene expression data for various diseases and biological systems. Dr. Liang will also direct the analysis of gene expression data and biomarker discovery and she will devote 10% of her effort to this project. Dr. Fang is a biostatistician and an assistant professor of Epidemiology and Preventive Medicine. His expertise is in the experimental design, statistical analysis for drug development, radiation biology and oncology, interval censored survival and multivariate survival analysis models. He will devote 10% effort. Dr. Dorsey will serve as key personnel for this Core to assist with microarray data normalization and preparation.
Showing the most recent 10 out of 21 publications
