- This event has passed.
Dr. David Tregouet, International Speaker Seminar Series
January 11, 2013
Please note change from usual room location.
CIHR STAGE International Speaker Seminar Series proudly presents
Dr. David Tregouet
Head director, Institute for Cardio-Metabolism and Nutrition (ICAN), Genomics Department, Pierre and Marie Curie University Campus
Research Director, UMRS937, Genomics of Venous Thrombosis, French National Institute of Health and Medical Research (“Inserm”)
Talk title: Haplotypes & Imputation, Two Complementary Tools: A Case Study on GenomeWide Expression Studies
David-Alexandre Trégouët was recruited at INSERM as a research associate in 1999 just after completing his PhD in statistical genetics. Since then, he has been working at INSERM UMR_S 937 (previously labelled UMR_S 525), a research department led by Francois Cambien, and focusing on cardiovascular genomics. In December 2009, he was promoted research director-2nd class and heads the team “Genomics of Venous Thrombosis” within INSERM UMR_S 937. In 2011, he was named director of the Genomics department of the Institute for Cardiometabolism and Nutrition (IHU-ICAN) and coordinator of the biostatistis/bioinformatics group of the post-genomics platform at the Pitié-Salpêtrière Hospital (P3S).
David-Alexandre Trégouët is recognized for his expertise in the analysis of genetic association studies and the development of statistical tools for the analysis of high-throughput microarray data as part of genome-wide association and expression studies (GWAS/GWES). He conducted the first two GWAS on venous thrombosis and developed tools for genome-wide haplotype association analysis. He belongs to the statistical core group of several past and current European projects including MORGAM, EURAGEDIC, Cardiogenics, CARDomics. He is the new investigator of the FARIVE study previously coordinated by Professor Joseph Emmerich and co-investigator of the MARTHA and MARFAST studies coordinated by Professor Pierre-Emmanuel Morange, these three studies focusing on venous thrombosis.
The development of high throughput technologies has stimulated comprehensive surveys on genome-wide expression and DNA variability data for disentangling the genetic architecture of human diseases. The genetics of transcript abundance has been extensively investigated through genome-wide expression studies (GWES). These studies demonstrated that, for a large fraction of genes (so-called eQTLs), expression is influenced by single nucleotide polymorphisms (SNPs) located in the vicinity of the regulated genes, generally referred to as cis eSNPs. The importance of cis eSNPs would be enhanced if they were associated at the same time with a disease, as such data would indicate that the associated gene is a candidate for the disease.
Most cis eQTL studies so far were based on single SNP analyses that did not account for the multiplicity of cis eSNPs that are often observed at an eQTL. One way to investigate whether associations observed at several cis eSNPs of the same eQTL are independent, or due to linkage disequilibrium (LD) between SNPs, is to conduct haplotype analysis, a strategy known to be able to distinguish “true” effect from those due to LD. Another approach is to perform GWES conditioning on the best cis eSNPs identified through a first run of GWES. The limitation of this strategy is that it is only able to identify cis eSNPs that have independent additive effects, contrary to haplotype analysis which can identify combinations of SNPs having non-additive effects or tagging a rare functional variant.
In this work, we conducted a systematic genome-wide search for haplotypic cis-acting effects on monocyte gene expression using data from the Cardiogenics Transcriptomic Study and replicate the main findings in the Gutenberg Healthy Study. Replicated findings were completed by 1000Genome based imputation analysis.
We identified 105 genes whose monocyte expression was under the influence of multiple cis-acting SNPs. About 75% of the detected genetic effects were related to independent additive SNP effects and the last quarter due to more complex haplotype effects. Of note, twenty-four of the genes identified to be affected by multiple cis eSNPs have been previously reported to reside at disease-associated loci. This could suggest that such multiple locus-specific genetic effects could contribute to the susceptibility to human diseases.