|
||||||||||||||||
HUM-MOLGEN -> Events -> Meetings and Conferences | ||||||||||||||||
Association Studies in Genomic Analysis | ||||||||||||||||
April 12, 2002 | ||||||||||||||||
Henry Stewart Conference Studies, The Cafe Royal; London Thursday 13th June 2002 CHAIRMAN'S INTRODUCTION: Dr Nick Short Managing Director and Chief Scientific Officer Iceland Genomics Corporation, Iceland KEYNOTE ADDRESS: GENETIC STUDIES IN DRUG AND DIAGNOSTICS DISCOVERY AND DEVELOPMENT - OPPORTUNITIES AND CHALLENGES Opportunity: confirmation of contributory role of candidate genes/polymorphisms to disease – not disease gene discovery Opportunity: refined (molecular) differential diagnosis – new diagnostics Opportunity: “pharmacogenetic” approaches to treatment Challenge: carrying out properly designed studies Challenge: biostatistical issues Challenge: public perception – managing expectations appropriately Professor Klaus Lindpaintner Vice President and Director, Roche Genetics, F. Hoffmann-La Roche AG, Switzerland THE CONSTRUCTION OF LINKAGE DISEQUILIBRIUM MAPS Linkage disequilibrium (LD) offers the prospect of high-resolution mapping of disease genes using high-density genotyping of single nucleotide polymorphisms (SNPs) in candidate regions. Statistical methods to analyse such data are in active development, but there has been relatively little attention given to the construction of LD maps. LD maps offer an immediate solution to the problem of optimal marker spacing and may provide more precise localisation of disease determinants. Such maps are analogous to the genetic linkage map to the extent that they represent the pattern of recombination in a region. Recent studies have shown a close relationship between recombination hot-spots and the transition regions between LD ‘blocks’. LD maps are expressed in linkage disequilibrium units (LDUs) which, if plotted against the physical location in kilobases, show a pattern of ‘plateaus’ and ‘steps’. The plateaus represent blocks which are regions of low haplotype diversity and therefore extensive disequilibrium and the steps represent recombination hot-spots. In this presentation, Dr Andrew Collins from the University of Southampton will present the methods used in the construction of LD maps and the results of the analyses in a number of genome regions. How are LD maps constructed in different populations likely to compare? Will the pattern of LD be preserved but the overall LD map lengths differ (reflecting different population histories)? To what extent will knowing the haplotype structure help when mapping a disease gene, particularly once there is already an LD map? Will it be possible to map a gene within a block or is that the limit of resolution by this method? Can LD maps and samples from different populations be profitably used to map within blocks? To what extent does the LD map mirror the linkage map? Is it reasonably representative of a high resolution (but sex averaged) linkage map? Is there any evidence for particular sequence motifs associated with the LD map ‘steps’? Dr Andrew Collins University of Southampton, UK BETTER MEDICINES THROUGH ASSOCIATION STUDIES - GETTING IT RIGHT There is little doubt that genetics will have a significant impact on the future management of common diseases such as type 2 diabetes, heart disease, asthma and osteoporosis. However, the vast majority of claims in the medical literature for discovery of common ‘disease genes’ are inconclusive, owing largely to poor study design leading in turn to lack of statistical power. Realising the potential of genetics relies on the demonstration of unequivocal gene effects through the execution of carefully designed experiments. Lessons from a real common disease ‘polygene’ The importance of careful study design The importance of having the correct clinical resources Moving from linkage to association When to believe a real association – making the effect ‘go away’ When is a ‘disease gene’ a disease gene? The role of functional validation Dr Simon Bennett Director of Business Development Oxagen Ltd, UK GAS IN PRACTICE: THE ICELANDIC CANCER PROJECT Association studies are widely recognised to be one of the most powerful techniques for isolating human alleles that predispose to disease, but their use raises serious problems. These include the large numbers of markers required, the need for very carefully matched controls, and the risk of false positives. Virtually all of these obstacles can be overcome by working in a suitable isolated population and using recently-available genetic information to define new marker sets. The Icelandic Cancer Project aims to take advantage of these opportunities to conduct a rigorous genome-wide scan for alleles predisposing to cancer. What is the ideal age for an isolated population? How big need an isolated population be to be useful, and how far does its usefulness increase with increasing size? How can genotyping be made more cost-effective? What is the most effective method of extracting haplotypes from genotyping data? (genome-wide association studies in cancer) Dr Nick Short Managing Director and Chief Scientific Officer Iceland Genomics Corporation, Iceland WHEN ARE ASSOCIATION STUDIES THE WAY TO GO AND WHEN ARE THEY NOT? Association studies require, by definition, substantial previous understanding of the role of particular genes, and are therefore not an ideal method for discovering basic new information about the links between genes and disease. In this presentation, Dr Kari Stefansson from deCODE genetics will describe how the uniquely comprehensive population data in Iceland and datamining products are used to gain new insights into the genetics of common, complex diseases. Once the key genes involved in disease have been identified, association studies both in Iceland and in other populations are used to increase the understanding of the role of these genes and the discoveries are applied in the creation of drug and diagnostic products for improving healthcare. Dr Kari Stefansson Chief Executive Officer deCODE genetics, Iceland SELECTING INFORMATIVE SNP SUBSETS FOR ASSOCIATION STUDIES Association studies of candidate genes involve the genotyping of single nucleotide polymorphisms (SNPs) throughout the gene of interest. If linkage disequilibrium (LD) exists betwen SNPs in the region, SNP genotypes are not independent, and only a subset of SNPs need to be genotyped for an informative association study. However, candidate genes may have many SNPs with a complex pattern of LD between them. This presentation will cover methods to determine informative subsets of SNPs, based on genotype data from a pilot study of less that 100 individuals. This can result in substantial reduction in genotyping for an association study, saving laboratory time and costs. How complex are the patterns of SNP LD to be found in candidate genes? How do LD measures summarise relationships between SNPs? What algorithms can be used to determine SNP subsets based on LD? How can haplotypes be used to ensure that SNP subsets capture all the genetic diversity within a region? What is the trade-off in study power versus numbers of SNPs to be genotyped? Dr Cathryn Lewis Guy’s, King’s and St. Thomas’ School of Medicine, UK EFFICIENT ASSOCIATION STUDIES USING POOLED DNA Systematic association analysis of complex human disorders is likely to require the genotyping of thousands of DNA polymorphisms on thousands of individuals. The analysis of pooled DNA samples has been proposed as a method of cost reduction that may make this feasible. Experimental errors in allele frequency estimation from pooled DNA Optimal design for pooled DNA studies on quantitative phenotypes Professor Pak Sham Institute of Psychiatry, London, UK COMPLEX AND MULTI FACTORIAL GENE DISCOVERY IN HUMAN PROPULATIONS: LESSONS FROM AIDS RESTRICTION GENES The application of population genetic principals and association mapping tools has led to the discovery and implication of eleven human genes that influence the come of exposure to and infection with HIV, the AIDS virus. The discoveries integrated advances from human molecular genetic genotyping technology, basic research on HIV pathogenesis to implicate candidate genes, exquisite curation of clinical data of AIDS cohort populations and computer algorithms that transform a century of population genetic theory and practice into a human gene mapping tool. The association discovered involved a learning curve for recognising true genetic signals and validating their efficacy by replication and interpretation of linkage disequilibrium patterning. The calculation of attributable risk demonstrates a primary role for discovered SNP variation in determining the landscape of epidemiological variation in this epidemic. Detecting and resolving genetic epidemiological signals in cohort studies Multi-disciplinary validation of population genetic distortion in disease categories Estimating the strength of the signal and its meaning to the disease Epistatic and gene interaction in association studies Candidate genes versus anonymous genome scans Ethnic group stratification - do we need to worry? Dr Stephen O’Brien Chief, Laboratory of Genomic Diversity Head of Section of Genetics National Institutes of Health (NIH), U.S. Registration and Coffee 09.00 Start of Proceedings 09.30 Close of Proceedings 17.30 |
||||||||||||||||
Organized by: | Henry Stewart Conference Studies | |||||||||||||||
Invited Speakers: | Dr Simon Bennett Director of Business Development Oxagen Ltd, UK Dr Andrew Collins University of Southampton, UK Dr Cathryn Lewis Guy’s, King’s and St. Thomas’ School of Medicine, UK Professor Klaus Lindpaintner Vice President and Director, Roche Genetics, F. Hoffmann-La Roche AG, Switzerland Dr Stephen O’Brien Chief, Laboratory of Genomic Diversity Head of Section of Genetics National Institutes of Health (NIH), U.S. Professor Pak Sham Institute of Psychiatry, London, UK Dr Nick Short Managing Director and Chief Scientific Officer Iceland Genomics Corporation, Iceland Dr Kari Stefansson Chief Executive Officer deCODE genetics, Iceland |
|||||||||||||||
Deadline for Abstracts: | N/A | |||||||||||||||
Registration: | Please visit our website at www.henrystewart.com for a registration form or contact: Susan Robinson, Conference Producer Tel: +44 (0)207 404 3040 Fax: +44 (0)207 404 2081 Email: susanr@henrystewart.co.uk |
|||||||||||||||
E-mail: | susanr@henrystewart.co.uk | |||||||||||||||
Posted by: | Susan Robinson | |||||||||||||||
Host: | no-dns-yet.demon.co.uk | |||||||||||||||
|
||||||||||||||||
Generated by meetings and positions 5.0 by Kai Garlipp |