A centralised facility for sequence to phenotype analyses

Genome-wide association studies (GWAS) have successfully identified hundreds of genes that determine the risk of common diseases. Despite the triumphs for many complex diseases, GWAS left a large part of the heritability unexplained. One of the explanations for this missing heritability concerns the contribution of rare genetic and structural variants which have not been captured in the current arrays used for GWAS but also not in the HAPMAP project which formed the basis for the imputations to characterize unmeasured genetic variants. To advance our understanding of human physiology and the pathogenesis of diseases, it is crucial to find these variants. In particular variants with modest to large effects on protein function are highly relevant for improving disease prediction and diagnostic models. Rare variants are captured with sequencing but are difficult to study in population-based studies. Rare variant research brings extended family studies back to the frontline of genetic research. First, sequencing can be done at lower intensity because sequence data from relatives can be used to correct inaccurate sequencing results and vice versa.Second, family studies can delineate recombination sites precisely. When sequencing is performed in a family-based study well-characterized for various phenotypes, sequencing data can be used to study hundreds phenotypes simultaneously. To identify relatively rare variants with modest to large effects on complex disease we aim: To build up a unique databank facility including multiple generation families who are characterized in depth phenotypically and genetically by exome sequencing Since many (but not all) rare variants with modest to large effects on protein function and disease risks are expected to be found in translated regions, i.e., exons of genes, we propose to sequence all exons of 800 persons from a family based study, the Erasmus Rucphen Family (ERF) study in order to evaluate the role of coding variations in disease. In the ERF study, we have documented over 1000 of endophenotypes. Endophenotypes are heritable risk factors for diseases that can be assessed as a quantitative outcome. Further, SNP data are available for the imputation of sequence data. The proposal submitted here focuses (but not limits itself) to two groups of endophenotypes (1) metabolic syndrome and related outcomes including body mass index, waist and hip circumference, blood glucose, CRP, adiponectin and cholesterol and phospholipids lipid levels as well as new phenotypes derived from metabolomics, (2) neuropsychiatric outcomes including migraine, ADHD, personality and personality disorder. These outcomes are major determinants of morbidity and mortality in Western societies. The biobank will be open to other researchers interested in genotype-phenotype studies in extended families.The data base will allow researchers in human and animal genetics to rapidly verify the evidence for a role of a gene in a certain phenotype to their interest. The data base can also be used for clinical applications such as the interpretation of the relevance of rare variants in personalized medicine.