Composite test for positive selection in the genome

Update: Must read post from p-ter.

A Composite of Multiple Signals Distinguishes Causal Variants in Regions of Positive Selection:

The human genome contains hundreds of regions whose patterns of genetic variation indicate recent positive natural selection, yet for most the underlying gene and the advantageous mutation remain unknown. We developed a method, Composite of Multiple Signals (CMS), that combines tests for multiple signals of selection and increases resolution by up to 100-fold. Applying CMS to candidate regions from the International Haplotype Map, we localized population-specific selective signals to 55 kb (median), identifying known and novel causal variants. CMS can identify not just individual loci but implicates precise variants selected by evolution.

From ScienceDaily:

Of the hundreds of these large genomic regions thought to be under positive natural selection in humans, only a handful have so far been winnowed to a precise genetic change. "Finding the specific genetic changes that are under selection can be like looking for a needle in a haystack," said Grossman.

Sabeti, Grossman and their colleagues wondered if there might be a way to enhance this genomic search. Because existing methods for detecting natural selection each measure distinct genomic features, the researchers predicted that an approach that combines them together could yield even better results.

After some initial simulations to test their new method, the research team applied it to more than 180 regions of the human genome that are thought to be under recent positive selection, yet in most cases, the specific gene or genetic variant under selection is unknown.

The researchers' method, called "Composite of Multiple Signals" or CMS, enabled them to dramatically narrow the size of the candidate regions, reducing them from an average of eight genes per region to one. Moreover the number of candidate genetic changes was reduced from thousands to just a handful, helping the researchers tease out the needles from the haystack.

...

In some cases, the researchers were able to identify a specific genetic change that is the likely focal point of natural selection. For example, a variation in a gene called protocadherin 15, which functions in sensory perception, including hearing and vision, appears to be under selection in some East Asian populations. Several other genes involved in sensory perception also appear to be under selection in Asia. In addition, the team uncovered strong evidence of selection in East Asians at a specific point within the leptin receptor gene, which is linked to blood pressure, body mass index and other important metabolic functions.

Citation: Grossman et al. A composite of multiple signals distinguishes causal variants in regions of positive selection. Science, 2010 DOI: 10.1126/science.1183863

More like this

Note: I'm introducing Do It Yourself as a new and hopefully semi-regular section on Genetic Future. The aim is to provide readers with instructions on how to access online resources for sequence analysis - an activity traditionally restricted to researchers, but one that will no doubt become more…
A few months ago I reviewed a paper which examined the various complexities of interpreting signals of natural selection from recently developed genomic tests in response to the avalanche of human sequence data. In the paper, Signals of recent positive selection in a worldwide sample of human…
A Genome-Wide Association Study Identifies Novel Alleles Associated with Hair Color and Skin Pigmentation: It has been a longstanding hypothesis that human pigmentation is tightly regulated by genetic variation. However, very few genes have been identified that contain common genetic variants…
Over the history of this weblog I have blogged about pigmentation a fair amount. The major reason is that that's where the money is; unlike height, let alone intelligence, the genetic architecture and evolutionary history of pigmentation has been elucidated with relative clarity. That is, we know…

Could I please beg of someone who has access to this paper to pass it on to me via email? My email address is, with "/" denoting a null char: Cauchy/Analytic/AT/Yahoo.com

Looking at eq. 1, it seems that this new method effectively counts an overlap between different signals? Little wonder there are few loci left. It's like the protein "interactomes" for mouse, worms and yeast that show basically no overlap.