Many researchers believe that rare genetic variants in humans might affect complex traits like lipoprotein levels, but that might not necessarily be the case at all. At this year’s American Society of Human Genetics meeting in San Francisco, CA, a team of scientists reported that most rare variants that they identified in a study of more than 67000 exomes have a weaker effect on complex traits than they had expected.
Although roughly 720,000 of these variants affected the resulting protein sequence, the team could only link a few of these variants to disease pathways. For example, the researchers found an association between a variant in the Apoc3 gene with lower triglyceride levels.
“There’s a big learning curve, and we’re just learning to analyze this data,” said team member Suzanne M. Leal, a professor of molecular and human genetics at Baylor College of Medicine and director of the college’s Center for Statistical Genetics.
Subjects were recruited from existing NHLBI studies, including the Jackson Heart Study, the Atherosclerosis Risk in Communities Study, and the Women’s Health Initiative. The team selected subjects based on disease phenotypes that they were interested in, like high- or low-lipid levels, extreme blood pressure, or early onset heart attacks. In all, the researchers analyzed more than 80 heart, lung, and blood phenotypes.
In contrast to genome-wide association studies that identify disease-associated variants scattered throughout the genome, this project only looked at protein-coding regions of the genome. Leal and her colleagues believe that this targeted approach has more potential to identify genetic links to disease.
Most of the variations were found in specific populations. For example, Darc gene variants were associated with varying white blood cell counts in African-Americans, while variants in the F7 gene appeared to influence levels of factor VII—a cog in the coagulation cascade—in European-Americans.
The study also confirmed some previously known associations, such as high levels of low-density lipoprotein—aka bad cholesterol—that track with certain variants of the apoB gene, and associations between LepR gene variants and levels of C-reactive protein, that increases the body's response to inflammation.
“One thing we saw was that the effect sizes (of rare variants) aren’t as great as many people predicted, so you need quite large sample sizes (to detect associations),” said Leal.
In some cases, the project provided leads that the researchers are now pursuing in follow-up studies. In these satellite studies, the researchers employed an exome chip—which was developed with direct input from the exome project—to analyze 240,000 coding variants.