How to use an article about genetic association? - Attia et al. - 2009 - Article


What are the goals of genetic studies?

There have been many publications on genetic associations of Alzheimer’s disease. Nowadays we use genetics in chronic diseases with major causes of morbidity and mortality. This is because it might shed light on the pathways that are involved in the diseases and identify targets for interventions. It can also help in improving the diagnostics or make sure that the therapeutic agents are more efficient and with less risk. This area is called pharmacogenomics. Genetics gives a short-term clinical effect, which is enhancing the risk stratification and providing information about prognosis.

Which genetic structures are important?

DNA has a double helix structure with strands on the sides, which are formed by alternating sugar and phosphate molecules. The rungs of the ladder contain four nitrogen-containing ring components, which are called bases. A base pair is always formed by adenine (A) and thymine (T) or by cytosine (C) and guanine (G). A nucleotide is a single base with its associated sugar and phosphate groups.

One chromosome is composed of one double-stranded DNA stretch. It contains many genes, with one gene being a stretch of DNA that codes for the building of one protein. We have 23 pairs of chromosomes, of which one pair is the sex chromosomes. This genome (entire DNA set) is formed of 23 chromosomes of the mother and 23 chromosomes of the father.

DNA is used to make proteins, that build cells and tissues and enzymes that catalyse certain biochemical reactions in a cell. This happens through two steps:

  1. transcription: DNA is transcribed into messenger RNA (mRNA). When mRNA migrates out the nucleus into the cytoplasm, it reaches the ribosome, which makes the proteins;

  2. translation: the mRNA is translated into proteins, that are made up of amino acids.

Through this process genetic information is converted, which is called the genotype. This forms the phenotype, together with other genes, proteins and environmental exposures.

What are DNA variations?

More than 99% of the DNA sequence is identical in all humans. However, we have so many base pairs, that still more than 12 million variations exist. The differences that don’t occur often (less than 1%) are called mutations. The differences that occur more frequently are called polymorphisms. Examples are:

  • the presence of absence of a DNA stretch. This involves DNA duplication, which is called copy number variation (CNV);

  • repeating patterns of DNA, with a variation in the amount of repetitions

  • a change in a single base pair, which is called single-nucleotide polymorphism or SNP. It is the most common polymorphism. There are two forms of SNPs in parts of genes that are translated: (1) non-synonymous SNPs change the amino acid sequence, and (2) synonymous SNPs don’t change the amino acids. Other SNPs occur in areas of the chromosome that don’t code for proteins, but can still influence the cell in other ways. It is usually indicated by ‘rs’.

A polymorphism can take different forms, called alleles. The APOE gene for instance has three alleles: e2, e3 and e4. The location of an allele is called a locus. The most common form of the gene is called the common allele and the others variant allele. The different alleles lead to different proteins, which are called isoforms.

Of the APOE gene, we mostly see the e3 allele in white populations. So e2 and e4 are called variant alleles. The apoE protein carries cholesterol and binds to a apoE receptor so that the cholesterol can be metabolised. The e2 isoform has a decreased strength of binding. The e3 and e4 isoforms have higher affinity.

Individuals get two APOE genes, one from the mother and one from the father. Any combination is possible: e2/e2, e3/e4, etc. When someone has two of the same alleles they are homozygous or homozygote. Two different alleles are called heterozygous or heterozygote.

When an individual has e2/e3 alleles, than they produce some e2 protein and some e3 protein. The protein that wins, is from the allele that is dominant. Of a dominant allele only one is necessary to carry all the biological activity. However when the allele is recessive there need to be two of them, otherwise they will remain biologically silent. If two differeing protein isoforms resulting from two different alleles share function, the model is called additive, or per-allele, model. The dominant, recessive and additive/per-allele models are models of inheritance or genetic models.

We see that people with a e2/e3 combination have part of the reduced affinity of the e2 allele and part of the higher affinity of the e3 allele. For most genetic association studies we don’t know the inheritance model.

How are genes distributed in populations?

We often want to know about the distribution of the alleles of interest in the population. Most allele distributions follow the Hardy-Weinberg equilibrium (HWE). These means that two alleles at a particular locus, A and a, with frequency p and q, result in genotype frequencies of AA, Aa and aa groups of p2, 2pq and q2 respectively, when there has been one generation of random mating.

Deviations from the HWE can be because of:

  • inbreeding, marrying close relatives, because the HWE is based on random mating;

  • genetic drift, which is a process in which a population is isolated and therefore there are limited options for mating;

  • migration;

  • very new mutations;

  • selection.

There could also be deviations in HWE because of methodological problems.

Candidate gene and genome-wide studies

Genetic association studies are often done on genes that are known through previous results or biology to be associated. Therefore this is called candidate gene association. Another way is screening the entire genome for associations. These genome-wide association (GWA) studies have resulted in many discoveries of genetic associations.

However, using GWA, can lead to the finding of many spurious associations, because you test so many potential genes at the same time. That is why the SNPs that seem to have a strong statistical signal are often tested for replication. These replication results are then published along with the initial data.

In genetic association studies the focus on population-based investigations in which the diseased and nondiseased individuals are unrelated. Other studies that can be carried out are linkage analysis. Here you research family members in large family trees, who are affected by a rare disease through a rare mutation.

What is meant by a 'linkage disequilibrium'?

With the genetic association studies we try to define causal relations. However we often find noncausal associations. To deal with these we use adjusted or multivariate analysis. This means that we isolate the function of a particular SNP from the other SNPs nearby in the gene. This is further complicated by linkage disequilibrium. This means that stretches of the genome are often inherited together. An association of a SNP with an outcome may therefore be noncausal, while the actual causal relationship is with another SNP. This distinction is important when we want to understand the underlying biology, but is not critical when using the SNP as a marker of risk.

“Haplotype blocks” are the result of linkage disequilibrium. These are DNA stretches defined by the presence of high linkage disequilibrium among the SNPs. Haplotypes are defined by two or more SNPs in linkage disequilibrium in the same haplotype block. The possible combinations of variants are dependent on the extent of linkage disequilibrium.

An example of this is the following: we have a common allele on SNP A with a frequency of 80% and a common allele on SNP B with a frequency of 60%. Without linkage allele A at SNP A and allele B at SNP B will occur together (0.80x0.60=) 36% of the time in the same person. With perfect linkage disequilibrium, it might be that allele A and B are always found together. The greater the measures, the greater the degree of linkage between the variants. A D’ of 1 indicates that the two alleles are always found together.

Page access
Public
Comments, Compliments & Kudos

Add new contribution

CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
Image CAPTCHA
Enter the characters shown in the image.