Genome wide association study

From The School of Biomedical Sciences Wiki
Jump to: navigation, search

A genome wide association study (GWAS) is a specific type of case-control study, which compares two groups of individuals by looking at their whole genome and assessing any genetic variants (the most common variant looked for is single nucleotide polymorphisms, SNPs). The case group contains individuals, which have a specific trait e.g. a disease, whereas the control group does not have the trait. By comparing a large sample size from each group, you can get an idea of any SNPs which are particularly prevalent in the case group and which are not in the control group. These SNPs show you possible sites of the genome, which may be important in the expression of the disease state and therefore once this study has been completed, more research can be done on the particular SNPs to find out more about the disease and possible ways of treatment.


GWAS can be analysed by plotting a Manhattan plot. This is a scatter graph which charts chromosome location on the x-axis and -log P value on the y-axis. The P value of something is the probability that an observed result has occurred by chance, therefore, a very low P-value means that the observation is more likely to be accurate. However, if we plotted the P-value on the Manhattan Plot, then the best results would be ones which the lowest values, so instead you can find the -log P value to reverse this. As a result, the most statistically significant results will show as the highest plots on the graph. Usually, a threshold value is set so as to separate the most significant results which should be further investigated from the lesser ones[1].


  1. Golan. D.E, Toshijian. A.H. (2012) Principles of Pharmacology, 3rd Edition, Lippincott Williams and Wilkins page. 78-9
Personal tools