Genetic methods for counting new species may be a little too good at their jobs, a new study suggests.
Computer programs that rely on genetic data alone split populations of organisms into five to 13 times as many species as actually exist, researchers report online January 30 in Proceedings of the National Academy of Sciences. These overestimates may muddy researchers’ views of how species evolve and undermine conservation efforts by claiming protections for species that don’t really exist, say computational evolutionary biologist Jeet Sukumaran and evolutionary biologist L. Lacey Knowles.
The lesson, says Knowles, “is that we shouldn’t use genetic data alone” to draw lines between species.
Scientists have historically used data about organisms’ ecological distribution, appearance and behavior to classify species. But the number of experts in taxonomy is dwindling, and researchers have turned increasingly to genetics to help them draw distinctions. Large genetic datasets and powerful computer programs can quickly sort out groups that have become or are in the process of becoming different species. That’s especially important in analyzing organisms for which scientists don’t have much ecological data, such as insects in remote locations or recently extinct organisms.
Knowles and Sukumaran, both of the University of Michigan in Ann Arbor, examined a commonly used computer analysis method, called multispecies coalescent, which picks out genetic differences among individuals that have arisen recently in evolutionary time. Such differences could indicate that a population of organisms is becoming a separate species. The researchers used a set of known species and tested the program’s ability to correctly predict…