Many estimates of how strongly traits and diseases share genetic signals may be inflated, according to a new UCLA-led study that suggests current methods for evaluating genetic relationships between traits fail to explain mating patterns.
Using powerful genome-sequencing technology, scientists have in recent years sought to understand genetic associations between traits and disease risk, in the hope that discoveries of shared genes will point clues to disease treatment. However, the UCLA researchers said their new study, which was published on November 17 Sciences, Provides caution against relying too much on genetic association estimates. They say such estimates are confounded by more non-biological factors than previously estimated.
Estimates of genetic relatedness usually assume that the mating is random. But in the real world, partners tend to pair up because of many shared interests and social structures. As a result, some of the genetic associations in previous work that have been attributed to co-biology may instead represent incorrect statistical assumptions. For example, previous estimates of genetic overlap between body mass index (BMI) and educational attainment likely reflect this type of demographics, caused by “assortative mating across traits,” or how individuals of one trait tend to partner with individuals of another trait. .
Estimates of genetic association deserve further scrutiny, the study authors said, as these estimates have been used to predict disease risk, gather evidence for potential treatments, inform diagnostic practices, and form arguments about human behavior and societal issues. The authors said that some in the scientific community have focused too much on estimates of genetic association based on the idea that studying genes, because they are immutable, can overcome confounding factors.
“If you look at two traits that are elevated in a group of people, you can’t conclude that they are there for the same reason,” said lead author Richard Border, a postdoctoral researcher in statistical genetics at UCLA. “But there was a kind of assumption that if you could trace this back to genes, you would have the causal story.”
Based on their analysis of two large databases of marital traits, the researchers found that assortative mating across traits correlated closely with estimates of genetic relatedness and plausibly represented a “significant” portion of estimates of genetic relatedness.
said study co-author Noah Zeitlin, professor of computational medicine and neuroscience at UCLA Health.
The researchers also examined genetic association estimates for psychiatric disorders, which have sparked controversy in the psychiatric community because they appear to show genetic relationships between disorders that appear to have few similarities, such as attention deficit hyperactivity disorder and schizophrenia. The researchers found that the genetic associations of a number of unrelated traits could reasonably be attributed to assortative mating across the traits and imperfect diagnostic practices. On the other hand, their analysis found stronger links for some trait pairs, such as anxiety disorders and major depression, suggesting at least some shared biology.
“But even when there is a real signal, we still suggest that we overestimate the extent of this involvement,” Border said.
Other study authors include Georgios Athanasiades, Alphonso Boyle, Andrew J. Schork, Na Kaye, Alexander I. Young, Thomas Werge, Jonathan Flint, Kenneth S. Kindler, Sriram Sankararaman, and Andy Dahl. The authors declared no conflicts of interest. Please see the study for a full list of funders.