University of Cambridge > Talks.cam > Microsoft Research Machine Learning and Perception Seminars > New probabilistic methods for inference of natural selection on regulatory sequences in the human genome

New probabilistic methods for inference of natural selection on regulatory sequences in the human genome

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Microsoft Research Cambridge Talks Admins.

This event may be recorded and made available internally or externally via http://research.microsoft.com. Microsoft will own the copyright of any recordings made. If you do not wish to have your image/voice recorded please consider this before attending

For decades, it has been hypothesized that gene regulation has played a central role in human evolution, yet much remains unknown about the genome-wide impact of regulatory mutations. Here we use complete genome sequence data to demonstrate that natural selection has exerted a profound influence on human regulatory sequences since our divergence from chimpanzees 4-6 million years ago. Our analysis is based on a new probabilistic method for characterizing the influence of natural selection on collections of short regulatory elements scattered across the genome. Our method, called Inference of Natural Selection from Interspersed Genomically coHerent elemenTs (INSIGHT), uses a generative probabilistic model to contrast patterns of genetic variation in humans and nonhuman primates in the elements of interest with those in nearby “neutral” sites. Using a Bayesian approach, we are able to pool weak information from many short elements in a manner that accounts for variation across the genome in patterns of neutral genetic variation. The model is efficiently fitted to genome-wide data by an approximate expectation maximization algorithm. Using simulations, we show that INSIGHT can accurately estimate the evolutionary parameters of interest even in complex evolutionary scenarios. We apply it to real genomic data and find that binding sites have experienced somewhat weaker selection than protein-coding genes, on average, but that the binding sites of several transcription factors show clear evidence of adaptation. We project that regulatory elements may make larger cumulative contributions than protein-coding genes to both adaptive substitutions and deleterious polymorphisms, which has important implications for human evolution and disease.

This talk is part of the Microsoft Research Machine Learning and Perception Seminars series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.

 

© 2006-2014 Talks.cam, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity