A visualisation of large-scale gene expression studies. Each dot represents a gene, and links between dots occur where there is a certain level of correlation between their expression patterns.

Genetic studies

What form do these studies take and how do they help us to pin down the role of the human genome’s 20,000 or so genes in different diseases? By Chrissie Giles

In 1902, Archibald Garrod showed that the inheritance of the disease alkaptonuria could be explained by Mendel’s recently rediscovered rules. By taking family history information and urine from his patients, he deduced that alkaptonuria was a recessive disorder, caused by an “inborn error” in a single gene. A century later, more than 6,000 single-gene disorders have been identified, ranging in frequency from cystic fibrosis – which occurs in about 1 in 3,000 people – to diseases that are so rare that only a few cases have ever been identified worldwide.

Studying families with rare genetic disorders remains a classic method of tracing the genetic basis of disease. If a large family group affected by a disease exists, researchers can hunt for genetic markers showing the same inheritance patterns as the disease and then, hopefully, track down the defective gene near the marker. To help with such studies, the National Autozygosity Mapping Resource was established in 2000 by clinical geneticists from across the UK. More than 3,000 blood relatives with autosomal recessive disorders (in whom both forms of the disease-causing gene are mutated) have been recruited.

Isolated populations – such as the Finns, Icelanders, Amish and Sardinians – can also be particularly useful in the search for the causes of genetic disorders. They tend to have a relatively uniform genetic background, a similar environment and little migration; some have detailed records of births and deaths dating back hundreds of years, allowing people’s ancestors to be traced, often back to the founders. If a population began with a few individuals, one or more of whom carried a particular genetic variation that causes disease, that variant may come to be represented in many of the descendants and the disease may be disproportionately frequent.

For example, Ellis–van Creveld syndrome – which causes short stature, polydactyly (extra fingers or toes), and problems with nails, teeth and heart – is far more common in the Amish population than in the rest of the USA. The mutated gene that causes the syndrome has been traced back to Samuel King and his wife in the mid-1700s, who passed the gene on to their descendants.

Complex questions

For common diseases such as heart disease and diabetes, the picture is far more complex. Genes undoubtedly play a part – along with environmental factors – but not in the simple ‘cause-and-effect’ way of single-gene disorders. Multiple genes are likely to be involved, with certain variants of each gene influencing the risk of disease by perhaps only a few per cent.

Studies of twins can help to reveal the influence of genes on a particular characteristic or disease. Comparing members of a non-identical twin pair to that of a pair of identical twins, raised together or apart, can produce an estimate of the ‘heritability’ of a trait. St Thomas’ Twin Research and Genetic Epidemiology Unit, King’s College London, holds one of the world’s largest twin databases and has published data on the degree of heritability of many different traits, from osteoarthritis (60 per cent heritable) to religious belief (40 per cent heritable) and number of sexual partners (38 per cent heritable).

Tracking down the genes involved can again take advantage of isolated populations. In Finland, for example, small groups of settlers moved north into uncolonised parts of the country during the 16th century, founding villages and towns that grew rapidly but received little immigration until World War II. Research by Professor Leena Peltonen-Palotie and colleagues at the University of Helsinki, Finland, has identified not only more than 30 single-gene diseases that are more frequent than in any other population but also genetic regions linked to several disorders, including multiple sclerosis, schizophrenia and autism.

Larger-scale studies look at a sample of the population as a whole. A classic case-control association study takes a group of people with a particular disease (‘cases’) and a group without the disease (‘controls’) and looks for genetic markers that are more or less common in one of the groups than in the other.

Some associations are very strong (such as those between certain variants of the human leukocyte antigen region and diseases such as ankylosing spondylitis and type 1 diabetes), leading to big differences in the frequency of the variant between cases and controls – but others with a smaller effect have proved difficult to pin down. Furthermore, if studies are too small, the statistics may not be reliable and the associations difficult to reproduce, and some variants may be relevant to a disease in one population but not in another. For the first time, advances in experimental technology – such as high-throughput genotyping methods that enable researchers to screen thousands of genetic variants at once – mean that association studies with enough power to detect even modest associations are now affordable.

The Wellcome Trust Case Control Consortium (WTCCC) is one of the biggest series of case-control studies to date. In the WTCCC, 24 teams of geneticists will compare the DNA of people with tuberculosis, coronary heart disease, type 1 diabetes, type 2 diabetes, rheumatoid arthritis, Crohn’s disease, bipolar disorder and hypertension (2,000 for each disease group) with that of the DNA from 3,000 healthy controls. The researchers will compare 500,000 single-nucleotide polymorphisms (DNA sequence variations) in individuals having a particular disease with those of individuals without the condition.

Matching cases and controls can be difficult because populations are not uniform but reflect the genetic history of hundreds or thousand of years of migrations and immigration. The Wellcome Trust, therefore, has funded an initiative led by Sir Walter Bodmer and colleagues at the University of Oxford, who are mapping variation in human genotypes across the UK. The People of the British Isles project is identifying individuals around the country whose families have been living in an area for several generations; volunteers are extensively genotyped. The project will help to reduce the likelihood of chance associations linked to population genetic substructure.

Longitudinal studies start with a disease-free study population and follow them over a period to see which illnesses develop. As data are being collected prospectively (that is, before disease onset – ideally from birth), it can be easier to establish which factors the subjects have been exposed to. For example, the Avon Longitudinal Study of Parents and Children (ALSPAC) – often called ‘Children of the 90s’ – has been studying more than 14,000 children and their parents in the former county of Avon for 15 years. Long-term studies also enable nested case-control studies: from a longitudinal study population, a subset of the population with a particular trait can be selected and compared with a subset without the trait.

Finding that a variant of a particular gene or genetic region is associated with a disease is by no means the end of the story, but it does open up new avenues for researchers to explore. The role of the gene in a disease pathway can be unravelled and new screens for associated genes created, often very quickly after the gene has been discovered. In the longer term, the new understanding of the disease mechanisms arising from genetic studies could lead to novel therapies and treatments.

A version of this article first appeared in ‘Wellcome Science’ (February 2007).

Lead image:

A visualisation of large-scale gene expression studies. Each dot represents a gene, and links between dots occur where there is a certain level of correlation between their expression patterns.

The Sanger Institute, Wellcome Images


Further reading

About this resource

This resource was first published in ‘Food and Diet’ in June 2011 and reviewed and updated in August 2016.

Genetics and genomics, Health, infection and disease
Food and Diet
Education levels:
16–19, Continuing professional development