Genetic links between traits are overstated, according to researchers
Estimating how strongly traits and diseases share genetic signals is often inflated. The new UCLA-led study shows how current methods for assessing genetic relationships between traits fail to account for mating patterns
Scientists hope that the discoveries of shared genetics will point to clues for tackling diseases through the use of powerful genome sequencing technology
According to UCLA researchers, the new study provides caution against relying too much on genetic correlation estimates Estimates are more complex by non-biological factors than they have been appreciated before
The estimates of genetic correlation are based on the assumptions of random sex. In the real world, people tend to get along with each other based on interests and social structures
Some genetic correlations in previous work that have been attributed to shared biology may instead reflect incorrect statistical assumptions
Most genetic correlation estimates assume that there is no chance at all. The social structure of the real world encourages people to work together to get the best results
Many of the same interests and values are shared by the people in the real world. To find out if there is any shared genetics between the two groups, previous work will be re-examined with more appropriate genetic models
Studies of the genetics of body mass index and educational achievement are likely to be influenced by the "cross-trait assortative mating" phenomenon
Estimates of genetic correlations deserve more scrutiny, since they have been used to predict disease risk, glean clues for potential therapies, and shape arguments about human behavior and societal issues
Some scientists argue that studies of genes, which are permanent, can overcome the effects of confounders
community has placed too much emphasis on using genetic correlations or the extent to which genes can be used to predict one another.
This research finds that cross-trait assortative mating is strongly associated with genetic correlations and plausibly accounts for a substantial portion of them
One of the researchers, Dr. Eric Dishon, an assistant professor of psychiatry at Penn State Hershey Medical Center, said he was not surprised by the findings
He said, however, it might lead to further study of mental health disorders in order to identify more specific genes for each disorder.
There is evidence of assortative mating for a number of genetically unrelated phenotypes, such as height and BMI
it is plausible that cross-trait assortative mating might be occurring, since the traits are diagnosable, but the imperfect nature of diagnostic measures may be responsible for the genetic correlationsI
On the other hand, their analysis found strong links between pairs of traits, like anxiety disorders and major depression, suggesting that there is some shared biology.