Evolutionary models accounting for multi-layer selection pressures (Visiting Scholar)

The evolutionary selection forces acting on a protein are commonly inferred using probabilistic codon models, by contrasting the rate of nonsynonymous to synonymous substitutions. Most of these models assume that the synonymous rate is homogenous across all sites, and thus it serves as a baseline to compare with the nonsynonymous rate. However, other and we have previously shown that synonymous substitution rates may vary substantially among sites, reflecting constraints acting at the mRNA and DNA levels. These observations led to the understanding that protein evolution is characterized by many layers of mutation and selection pressures. I intend to develop novel evolutionary models that distinguish between mutation pressure, selection at the mRNA and DNA level, and selection at the amino-acid level. These models will be tested on a large set of protein coding genes as well as by extensive simulation studies. Our preliminary results show that the suggested models fit vertebrate coding sequences significantly better than commonly used codon models, in which the multi-layer selection forces are ignored. Furthermore, a preliminary genomic screen shows that for vertebrates, accounting for variability of selection at the DNA and mRNA level substantially reduces the number of genes predicted to evolve under positive selection. The novel methods suggested in this proposal should provide a robust approach to detect extremely conserved regions within protein coding genes, as well as to infer sites and lineages experiencing adaptive evolution, a central task in today's genomic analyses.

Browse all 2 Items

Evolutionary models accounting for multi-layer selection pressures (Visiting Scholar)

Project

Title:
Evolutionary models accounting for multi-layer selection pressures (Visiting Scholar)
Permalink:
https://idn.duke.edu/ark:/87924/r45947
Temporal:
2010-2011
Creator:
Pupko, Tal
Type:
Collection
Description:
Project
Abstract:
The evolutionary selection forces acting on a protein are commonly inferred using probabilistic codon models, by contrasting the rate of nonsynonymous to synonymous substitutions. Most of these models assume that the synonymous rate is homogenous across all sites, and thus it serves as a baseline to compare with the nonsynonymous rate. However, other and we have previously shown that synonymous substitution rates may vary substantially among sites, reflecting constraints acting at the mRNA and DNA levels. These observations led to the understanding that protein evolution is characterized by many layers of mutation and selection pressures. I intend to develop novel evolutionary models that distinguish between mutation pressure, selection at the mRNA and DNA level, and selection at the amino-acid level. These models will be tested on a large set of protein coding genes as well as by extensive simulation studies. Our preliminary results show that the suggested models fit vertebrate coding sequences significantly better than commonly used codon models, in which the multi-layer selection forces are ignored. Furthermore, a preliminary genomic screen shows that for vertebrates, accounting for variability of selection at the DNA and mRNA level substantially reduces the number of genes predicted to evolve under positive selection. The novel methods suggested in this proposal should provide a robust approach to detect extremely conserved regions within protein coding genes, as well as to infer sites and lineages experiencing adaptive evolution, a central task in today's genomic analyses.
Identifier:
  • 223
  • duke:277089
  • ark:/87924/r45947
  • 746f5c9c-9d82-41e1-b384-08327c142f0b