Gene Algorithm May Be Able to Predict If You're Gay
POTENTIAL FOR MISUSE TROUBLES LEAD SCIENTIST, WHO'S JUST LEFT THE LAB
By Elizabeth Armstrong Moore, Newser Staff
newser.com
Posted Oct 10, 2015 8:16 AM CDT
(NEWSER) – Scientists have developed an algorithm that uses epigenetic data from nine regions scattered across the human genome to predict male sexual orientation with 70% accuracy. But after lead author Tuck Ngun at the University of California, Los Angeles, presented his team's findings at the American Society of Human Genetics annual meeting in Baltimore this week, critics said the test is prone to false positives due to, as one researcher puts it, "spurious associations," reports New Scientist. Another added that it needs to be replicated "on larger samples in order to know how good it is, but in theory it's quite interesting."
To devise the machine learning algorithm they call FuzzyForest, Ngun's team studied 37 pairs of identical twins with different sexual orientation and 10 with the same, the genetics society says in a press release. The test seems doomed either way, getting criticism for not being accurate enough while alarming others, including Ngun, for its potential if accurate to be misused. "I'm gay," Ngun tells the New Scientist, "and I've always wondered why I am the way I am. But once you have this information, you can't control how it's used or disseminated." It's possible, for instance, for the test to become a fetal screening tool. Ngun says he has actually left his lab this past week, abandoning the research altogether because of "the potential for misuse of the information." (Almost one in three young Americans say they are some degree of bisexual.) |
