Statistical Genomics - Brooke L. Fridley

Statistical methods research

• Bayesian methods for genetic and genomic association studies

  • Bayesian hierarchical nonlinear models for pharmacogenomic studies. Funded by the National Institutes of Health. (R21 GM 86689)

    The goal of this grant is to develop bayesian hierarchical nonlinear methods to model the relationship between cytotoxicity endpoints and genomic data for cancer pharmacogenomic studies. Dr. Fridley's Statistical Genomics lab is also extending its models to incorporate gene and pathway information in order to assess gene and pathway effects on cytotoxicity.

• Methods for integration of multiple types of genomic data into association studies

  • Integrative genomic models for analysis of pharmacogenomic studies. Funded by the National Cancer Institute. (R21 CA 140879) (completed)

    The goal of this research was to combine various sources of phenotypic and genomic data collected into pharmacogenomic models using bayesian methods. To accomplish this, our lab combined the ideas of path analysis and bayesian hierarchical model with stochastic search variable selection.

• Gene set and pathway analysis methods for mRNA expression and genetic association studies

  • Determination of novel gene set and pathways involved in ovarian cancer risk and survival. Mayo Clinic SPORE in Ovarian Cancer Pilot Project Award. Funded by the National Cancer Institute. (P50 CA 136393) (completed)

    This study sought to determine the optimal approach for conducting gene set analysis for genomewide association studies, followed by application of this approach to a study involving ovarian cancer.

• Statistical and bioinformatics analysis methods for association analysis involving sequence data.