Department of Genetics
Department of Biostatistics
Carolina Center for Genome Sciences
Lineberger Comprehensive Cancer Center
Curriculum in Genetics and Molecular Biology
Curriculum in Bioinformatics and Computational Biology
Ph.D., University of Michigan, Ann Arbor, Biostatistics, 1996-2001 Advisor: Michael Boehnke, PhD
M.S., University of Michigan, Ann Arbor, Biostatistics, 1994-1997
M.A., University of California, Los Angeles, Mathematics, 1991-1994
B.S., University of California, Los Angeles, Applied Mathematics, 1986-1990
Keywords: complex disease models, statistical genetics
My research interests focus on the development and application of statistical methods for mapping complex trait and disease susceptibility genes. My primary areas of application have been prostate cancer and cardiovascular disease.
I started working with my closest collaborator Dr. Kathleen Cooney, an oncologist, on the then newly formed University of Michigan Prostate Cancer Genetics Project (UM-PCGP) in 1995 when I was a graduate student at the University of Michigan. At the time I had several years of experience with linkage analyses and linkage disequilibrium mapping that localized the gene responsible for ataxia-telangiectasia. In 2003, Dr. Cooney and I used linkage analysis to map a prostate cancer susceptibility gene to a region on chromosome 17q21-22. In collaboration with investigators at John Hopkins and Wake Forest Universities we subsequently fine-mapped the linkage interval in 2007 and just recently sequenced 94 unrelated subjects in our strongest families across all gene-coding regions inside the linkage candidate region. Through this effort, we identified a recurrent mutation in the gene HOXB13 – a strong candidate based on HOXB13’s biological role in prostate development. We subsequently genotyped the identified mutation in ~5,000 additional unrelated prostate cancer cases enriched for positive family history and early-onset disease and 1400 screened controls. We found strong evidence for an association between this mutation and prostate cancer and found a significantly increased frequency of the mutation in prostate cancer cases with positive family history and early-onset disease compared to men absent family history diagnosed with prostate cancer later in life. This finding was recently published in the New England Journal of Medicine and represents the first uncommon high-penetrant mutation identified for prostate cancer. HOXB13 mutation screening should become a valuable tool for future prostate cancer screening studies. We have a number of ongoing studies including work to further characterize the role of HOXB13 mutations in prostate cancer and expanding our search for uncommon coding variants associated with the disease. We are additionally in the process of finalizing the first stage of a two-stage genome-wide association study for common genetic variants that are associated with early-set forms of the disease.
In addition to my work with prostate cancer, I have collaborated on many genetic studies for a wide-range of quantitative traits and diseases including obesity, cardiovascular disease, type 2 diabetes, cystic fibrosis, HIV and asthma. Many of these studies have involved analyses of common genetic variants through genome-wide association studies. Recent technological advances in “next-generation” high-throughput sequencing have created exciting new opportunities to identify uncommon high-penetrant mutations like those that we found for prostate cancer. I am an investigator on several NIH funded studies, including the NHLBI funded Exome Sequencing Project and a large whole-genome sequencing project to identify uncommon genetic variants that are associated with cannabis and stimulant dependence. This latter project is headed by UNC Department of Genetics colleague Kirk Wilhelmsen. With respect to sequence data, I have a particularly strong interest in studying under-represented minority populations and subjects with “extreme” phenotypes – i.e. subjects with highly unusual quantitative measures such as extreme high and low LDL measures or case-control studies based on subjects with early-onset disease and/or strong positive family history.
In addition to my work on collaborative gene-mapping studies, I enjoy developing and applying new methods to genetic data. My methodological research has included developing novel approaches for accurately estimating statistical power in genetic association studies, designing haplotype-based association analyses using extended haplotype sharing, creating powerful nonparametric linkage statistics and using prior hypotheses and applying the false discovery rate to subsets of data to increase power of gene-trait association discoveries. I have a particularly strong interest in optimizing study designs by including existing publically available data along with investigator-collected data. Our two-stage genome-wide association study for early-onset prostate cancer is based on a method my former student, Lindsey Ho, and I devised based on using freely available genotype data on thousands of unscreened controls in Stage 1 and genotyping our own screened study controls in Stage 2. We calculate the optimal proportions of our early-onset prostate cancer cases to be divided into the two stages and use a replication-based design where the results in Stage 1 reduce the multiple test burden in Stage 2. Lindsey and I showed in our recently published manuscript in Human Genetics, that this kind of strategy maintains valid hypothesis tests, and results in increased statistical power and significantly decreased total costs compared to studies that do not include free public control genotype data.
Ewing CM*, Ray AM*, Lange EM*, Zuhlke KA, Robbins CM, Tembe WD, Wiley KE,..., Carpten JD, Isaacs WB, Cooney KA. (2012) Germline mutations in HOXB13 are associated with prostate cancer risk. New England Journal of Medicine (in press).
Croteau-Chonka DC, Wu Y, Li Y, Fogarty MP, Lange LA, Kuzawa CW, McDade TW, ..., Adair LS, Lange EM, Mohlke KL. (2012) Population-specific coding variant underlies genome-wide association with adiponectin level. Human Molecular Genetics 21: 463-471.
Lange EM, Salinas CA, Zuhlke KA, Ray AM, Wang Y, Lu Y, Ho LA, Luo J, Cooney KA (2012) Early onset prostate cancer has a significant genetic component. The Prostate. 72: 147-156.
Wang Y, Ray AM, Johnson EK, Zuhlke KA, Cooney KA, Lange EM (2011) Evidence for an association between prostate cancer and chromosome 8q24 and 10q11 genetic variants in African American men: The Flint Men’s Health Study. The Prostate 71: 2225-231.
Ho LA, Lange EM (2010) Using public control genotyping data to increase power and decrease cost of case-control genetic association studies. Human Genetics 128:597- 608.
Joubert BR, North KE, Wang Y, Mwapasa V, Franceschini N, Meshnick SR, Lange EM (2010) Comparison of genome-wide variation between Malawians and African ancestry HapMap populations. Journal of Human Genetics 55:366-374.