Experts have developed a new statistical tool which is more accurate at determining causal genes and variants for a disease, and reducing the number of false positives.
The new method, causal-Transcriptome-wide Association studies (cTWAS), uses advanced statistical techniques, pooling data from genome wide association studies (GWAS) and predictions of genetic expression.
GWAS is a common method used to try to pinpoint genes linked to different diseases. As an example, it identifies the genome sequences of large numbers of people with a specific disease compared with those from people who are disease free.
The differences seen in those with disease could highlight the difference in genes that could raise the risk for that particular disease, triggering further research.
- Researchers unveil how genes can impact body’s response to nutrients
- Rare genetic variants that contribute to male hair loss identified in study
- Human cells grown in artificial eye could transform lives of people with macular degeneration
The difficulty lies in the fact that most diseases are not the result of a single genetic variation, rather they are caused by the complex interplay between several genes, environmental factors, and lots of other variables.
This means that GWAS can highlight a number of variants across many regions in the genome that are linked to a specific disease. It does not, however, identify causality.
An occurrence called linkage disequilibrium results in many variants across an average genomic region to be highly correlated with each other. This is due to the fact that DNA is passed down in blocks, rather than individual genes.
The study’s senior author, Dr Xin He, Associate Professor of Human Genetics at the University of Chicago, explained: “You may have many genetic variants in a block that are all correlated with disease risk, but you don’t know which one is actually the causal variant.
“That’s the fundamental challenge of GWAS, that is, how we go from association to causality.”
Gene expression levels are used to try to overcome these challenges. Expression quantitative trait loci (eQTLs) are genetic variants linked to gene expression.
eQTL data is used on the basis that if a variant linked with a disease is an eQTL of some gene A, then A could be the link between the variant and the disease.
Trouble arises when nearby variants and eQTLs of different genes can be associated with the eQTL of gene A while affecting the disease directly. This results in a false positive.
Current methods lead to false positive genes more than half of the time.
- Critical genes for type 1 diabetes identified for first time
- Obesity genes can trigger and prevent type 2 diabetes
- Genetic link between blood sugar levels and migraine confirmed
The issue led to the development of a new method, cTWAS, that reduces the number of false positives by using advanced statistical techniques.
It has been created by Professor He and Dr Matthew Stephens, the Ralph W. Gerard Professor and Chair of the Departments of Statistics and Professor of Human Genetics.
This latest method takes into account multiple genes and variants rather than just concentrating on a gene at a time. It removes genes and variants that muddy the water.
Professor He said: “If you look at one at a time, you’ll have false positives, but if you look at all the nearby genes and variants together, you are much more likely to find the causal gene.
“The software will allow people to do analyses that connect genetic variations to phenotypes. That’s really the key challenge facing the entire field. We now have a much better tool to make those connections.”
