In order to bring you the best possible user experience, this site uses Javascript. If you are seeing this message, it is likely that the Javascript option in your browser is disabled. For optimal viewing of this site, please ensure that Javascript is enabled for your browser.
Did you know that your browser is out of date? To get the best experience using our website we recommend that you upgrade to a newer version. Learn more.

We use cookies to optimise the design of this website and make continuous improvement. By continuing your visit, you consent to the use of cookies. Learn more

Genetic determinants of cardiovascular disease: are genome-wide association studies the whole answer?

  • Genome-wide association studies: only a partial success story? Presented by H Watkins (Oxford, GB) - Slides
  • Where to find the missing heritability? Presented by F Cambien (Paris, FR) - Slides
  • Is exon sequencing the answer? Presented by S Kaab (Munich, DE)
  • How to interpret sequence data, presented by D F Gudbjartsson (Reykjavik, IS) - Slides

Advances in genotyping technology have allowed for an exponential increase in genome-wide association studies (GWAS) and a complete change of the landscape of human genetic research over the last 5 years. This well attended session chaired by Prof. Samani (Leicester) and Prof. Kääb (Munich) critically reviewed the current achievements and demonstrated limitations as well as perspectives for genetic research in common cardiovascular diseases.

Genome-wide association studies: only a partial success story?

Hugh Watkins (Oxford) reviewed and praised the unparalleled amount of entirely new information gained in the cardiovascular field by GWAS. Currently more than 30 independent genetic loci have been identified to be associated with coronary heart disease (CAD). Success in the past has arisen from studying rather crude endpoints in huge numbers rather than from sub-dividing disease entities. Most variants act via modest changes in gene expression. Same loci correspond to already existing drug targets giving hope that additional loci will be informative, identifying novel potential drug targets as well as novel signalling pathways. The nature of these large scale experiments requires multi-national collaborations and results in deposits of a wealth of open access data. While the benefit of GWAS for identifying potential drug targets and improved or novel disease pathways is undisputed, initial expectations on their utility for risk prediction need to be downgraded. Risk-prediction by genetic loci may have a greater role where prediction is limited with conventional measures or risk markers.

Where to find the missing heritability?

Francois Cambien (Paris) addressed the problem of missing heritability in the context of current GWAS. The number of firmly replicated trait-associated loci identified in GWAS is impressive (, but with rare exceptions, the contribution of individual single nucleotide polymorphisms (SNPs) to the studied phenotype is weak. The heritability quantifies how much of the variation of a trait in a population is explained by genetic variation. The following reasons why the fraction of heritability of common traits and diseases explained by the identified loci is small have been discussed:

  • Heritability may be over-estimated by family based studies
  • Disease-associated markers in GWAS are in most instances not directly responsible for the associations observed
  • GWAS cannot account for complex interactions
  • Only a fraction of the true associations are detected as statistically significant
  • Genotyping arrays are designed to evaluate associations with common variants, but could rare variants account for a large fraction of the heritability of complex traits?
  • The genetic contribution to complex traits implicates a very large number of loci with very weak effects.

In conclusion, the genetic framework that underlies complex traits involves a considerable number of variants covering a wide range of frequencies and having for the vast majority of them very weak effects. On the other hand, in human genetics, major breakthroughs with important medical implications frequently originated from the discovery of rare variants contributing almost nothing to disease heritability in the population.

Is exon sequencing the answer?

Stefan Kääb (Munich) substituted at short notice for Mark Lathrop (Evry) who was stranded in New York due to the current storm. He reviewed the potential of exon sequencing to address one of GWAS’ limitations, namely the investigation of uncommon genetic variants with allele frequencies below 2% and expected moderate effects of OR>2. These variants constitute a vast reservoir of genetic variations and may elucidate some of the missing or hidden heritability in common diseases.
Exome sequencing will provide a large number of synonymous as well as non-synonymous variants in any given individual requiring an adequate filtering system to allow for meaningful interpretation. Usually synonymous variants are being discarded (they could still be the cause of splice variants and other deleterious effects could be behind them) resulting in roughly 10.000 non-synonymous variants, one in every second gene on average. Exome sequencing is a powerful tool in identifying causal genes in rare diseases. An example of a gene identification in a autosomal recessive disorder, of a mitochondriopathy (complex I deficiency) in a single patient using exome sequencing identifying ACAD9 as the causal gene was presented (Haack et al. Nat Genet 2010). In association studies, to achieve adequate power, large samples sizes (n>10 000), large effect sizes (OR>2) plus additional functional evidence is required. Such functional evidence is usually easier to obtain for coding sequences than it is for non-coding ones. Large datasets are expected to be released in the near future.

How to interpret sequence data

Daniel Gudbjartsson (deCode genetics, Reykjavik) expanded the discussion on the potential of next generation sequencing. Uneven coverage of exons, the difficulties in detecting deletions and duplications and the additional cost of capturing constitute serious disadvantages for exome sequencing in the long run. Full genome sequencing methods are currently informative enough to identify and genotype most sequence variations. Exceptions include insertions, translocations and variants in problematic regions. Additional power can be gained by predicting function based on bioinformatics methods. In conclusion the current problems focus rather on variant interpretation than data acquisition.




Genetic determinants of cardiovascular disease: are genome-wide association studies the whole answer?

The content of this article reflects the personal opinion of the author/s and is not necessarily the official position of the European Society of Cardiology.