Meta-analysis of genome-wide linkage searches: extending the GSMA

Lead Research Organisation: King's College London
Department Name: Genetics and Molecular Medicine

Abstract

Finding genes that contribute to human diseases will enable us to identify individuals who are at increased risk of developing these diseases, and provide novel targets for drug development. However, disease risk is determined by the cumulative effect of many genes (10? 50? 100?) and identifying these genes has proved challenging. One common tool is linkage studies, where researchers study the pattern of DNA inheritance in families with at least two individuals affected with the disease. These studies should highlight chromosomal regions which contain disease genes. Many linkage studies have been performed, into many different diseases, by research groups in the UK and worldwide. However, the results have been uniformly disappointing, with each study identifying only few interesting regions, and little correlation between results within a disease. This limited success probably occurs because each gene causes only a slight increase in disease risk, so studies based on thousands of families are required, which is outside the scope of individual research centres. We have developed a statistical method of meta-analysis that can pool the results of different linkage studies, so the results of several small studies are synthesized into one large, informative study. This method has previously been applied to diseases such as schizophrenia, coronary heart disease and inflammatory bowel disease, in each case identifying novel regions which are now the focus of further genetic studies. We now plan to extend the methodology behind this meta-analysis. The improved methodology will increase the number of target regions identified, ensure that each region has a high probability of containing a disease gene, and will provide narrower target intervals, thus reducing the amount of follow-up work required. We will apply these newly developed methods to (1) schizophrenia, (2) autoimmune disease (e.g. rheumatoid arthritis, psoriasis, type 1 diabetes), searching for genes that contribute to all of these diseases, and (3) obesity. Meta-analysis is a valuable tool in gene-hunting, as it provides a starting point for the follow-up studies that will identify the true genetic mutation. The meta-analysis of linkage studies is a cost-effective, efficient, and powerful method to localise genes. It provides excellent value-for-money by using previously generated, but disappointing, data to identify new leads for finding disease genes for the many diseases that cause a major public health burden in the UK and worldwide.

Technical Summary

Genome-wide linkage studies have been used extensively to identify regions which may harbour susceptibility genes for complex diseases but results have been universally disappointing. Few significant linkage signals have been identified, with little concordance between studies in the same disease. The Genome Search Meta-analysis method was developed by us to pool results across studies. Its simple design assesses the rank of the maximum linkage statistics (LOD score, NPL, p-values) reached in fixed chromosomal bins. This has been widely applied, but the current methodology has many limitations. In this application, we aim to extend the GSMA method by (1) incorporating a novel semi-parametric test statistic that retains some information of the magnitude of original LOD scores, (2) implementing a test for heterogeneity of linkage information across studies, and (3) providing false discovery rates based on genome-wide assessment of linkage in the GSMA. Most importantly, we will develop a fine-scale meta-analysis method based on dense multipoint linkage statistics, to improve the current coarse localisation information of the GSMA. This quantitative approach will combine LOD scores from each study at each chromosomal point, using time series modeling where necessary. These developments will improve the power of the GSMA to detect linkage; they will explore the heterogeneity of linkage evidence across studies, and provide improved localization of linkage evidence to facilitate follow-up studies. In addition, the GSMA will be applied to several disease areas. We will update our previous schizophrenia meta-analysis, applying the novel developments described above. A meta-analysis of linkage scans in all autoimmune diseases will be performed, to test the hypothesis that common genes predispose to these traits. We will also analyse traits related to obesity and body mass index; this will extend the GSMA to quantitative traits, developing methods for diverse ascertainment of families and phenotypes.

Publications


10 25 50
Bouzigon E (2010) Meta-analysis of 20 genome-wide linkage studies evidenced new regions linked to asthma and atopy. in European journal of human genetics : EJHG
Forabosco P (2009) Meta-analysis of genome-wide linkage studies across autoimmune diseases. in European journal of human genetics : EJHG
Hermanowski J (2007) Meta-analysis of genome-wide linkage studies for multiple sclerosis, using an extended GSMA method. in European journal of human genetics : EJHG
Zhou K (2008) Meta-analysis of genome-wide linkage scans of attention deficit hyperactivity disorder. in American journal of medical genetics. Part B, Neuropsychiatric genetics : the official publication of the International Society of Psychiatric Genetics
 
Description ADHD 
Organisation King's College London (KCL)
Department Institute of Psychiatry (IoP)
Country United Kingdom of Great Britain & Northern Ireland (UK) 
Sector Academic/University 
PI Contribution Contributed expertise in the design and analysis of meta-analysis of genome-wide linkage studies of attention deficit hyperactivity disorder
Collaborator Contribution Enabled us to particate in meta-analysis of linkage studies of ADHD, and explore how study heterogeneity affected behaviour of the method.
Impact 18988193
Start Year 2007
 
Description Alzheimer 
Organisation King's College London (KCL)
Department Institute of Psychiatry (IoP)
Country United Kingdom of Great Britain & Northern Ireland (UK) 
Sector Academic/University 
PI Contribution Contributed to study design and statistical analysis.
Collaborator Contribution Collaboration on meta-analysis of Alzheimer disease enabled us to develop statistical methods for dealing with genetic heterogeneity and incomplete data availability
Impact 19362756
Start Year 2007
 
Description Celiac disease 
Organisation School of Medicine
Department Department of Epidemiology
Country United States of America 
Sector Academic/University 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Celiac disease 
Organisation University Paris Sud
Country France, French Republic 
Sector Academic/University 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Celiac disease 
Organisation University of Gothenburg
Country Sweden, Kingdom of 
Sector Academic/University 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Celiac disease 
Organisation University of Helsinki (Helsingin yliopisto)
Country Finland, Republic of 
Sector Academic/University 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Celiac disease 
Organisation University of London
Department Institute of Cancer Research UK (ICR)
Country United Kingdom of Great Britain & Northern Ireland (UK) 
Sector Academic/University 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Celiac disease 
Organisation University of Utrecht
Department University Medical Center
Country Netherlands, Kingdom of the 
Sector Hospitals 
PI Contribution Instigated, co-ordinated, analysed and published meta-analysis of linkage studies in celiac disease, obtaining results from collaborators from their studies.
Collaborator Contribution Contributed results from genome-wide linkage study in celiac disease Contributed results from genome-wide linkage study in celiac disease for meta-analysis Contributed results from genome-wide linkage study in celiac disease for meta-analysisContributed results from genome-wide linkage study in celiac disease for meta-analysisContributed expertise on genetic studies of celiac disease
Impact 19622889
Start Year 2007
 
Description Depression linkage 
Organisation School of Medicine Stanford
Department Department of Psychiatry and Behavioral Sciences
Country United States of America 
Sector Academic/University 
PI Contribution Will perform meta-analysis of three genetic linkage studies of depression when primary analysis is complete
Collaborator Contribution Provided data for meta-analysis of linkage studies in depression
Impact None as yet - collaboration is on-going Multi-disciplinary - psychiatry, genetics and statistics
Start Year 2008
 
Description Depression linkage 
Organisation School of Medicine Utah
Department Divisiont of Genetic Epidemiology
Country United States of America 
Sector Academic/University 
PI Contribution Will perform meta-analysis of three genetic linkage studies of depression when primary analysis is complete
Collaborator Contribution Provided data for meta-analysis of linkage studies in depression
Impact None as yet - collaboration is on-going Multi-disciplinary - psychiatry, genetics and statistics
Start Year 2008
 
Description SLE linkage meta-analysis 
Organisation University of California
Department University of California, San Francisco
Country United States of America 
Sector Academic/University 
PI Contribution We performed the statistical analysis of meta-analysis of linkage studies of systemic lupus erythematosus, collaborating with UCSF to design study and analyse the data.
Collaborator Contribution Contacted collaborators to contribute genome-wide linkage study results for meta-analysis of linkage studies in systemic lupus erythematosus, contributed her expert knowledge of the phenotype, and worked with us on the data analysis.
Impact 16971955
 
Description Schizophrenia linkage meta-analysis 
Organisation Stanford University
Department School of Medicine
Country United States of America 
Sector Academic/University 
PI Contribution Formed consortium to perform meta-anlaysis of genome-wide linkage studies in schizophrenia. Requested, collaated and analysed results using our GSMA method.
Collaborator Contribution Collaborated closely on the meta-analysis of schizophrenia linkage analysis, contacting data providers to contribute their study results to the meta-analysis.
Impact 19349958