Transcript ppt
Lecture 22 : Phylogeography and Coalescence November 13, 2015 Final Exam We will have the exam on the last day of class, Monday, December 7 at 2:30 pm Exam will be in the computer lab, 3306 LSB Exam will last 50 minutes, so it will be half as long as the previous two exams. Non-cumulative, covers Mutation through Association Genetics Review session on Friday, Dec 4 Last Time Signatures of selection Hudson-Kreitman-Aguade Test Synonymous versus Nonsynonymous substitutions McDonald-Kreitman Molecular clock Introduction to phylogenetics Today Phylogeography Limitations of phylogenetic analysis Coalescence introduction Influence of demography on coalescence time Coalescence and human origins Choosing Phylogenetic Trees If multiple trees equally likely, select majority rule or consensus Strict consensus is most conservative approach Bootstrap data matrix (sample with replacement) to determine robustness of nodes E 60 Lowe, Harris, and Ashton 2004 A D F CB 60 60 Felsenstein 2004 Phylogeography The study of evolutionary relationships among individuals based on phylogenetic analysis of DNA sequences in geographic context Can be used to infer evolutionary history of populations Migrations Population subdivisions Bottlenecks/Founder Effects Can provide insights on current relationships among populations Connectedness of populations Effects of landscape features on gene flow Phylogeography Topology of tree provides clues about evolutionary and ecological history of a set of populations Dispersal creates poor correspondence between geography and tree topology Vicariance (division of populations preventing gene flow among subpopulations) results in neat mapping of geography onto haplotypes Example: Pocket gophers (Geomys pinetis) Fossorial rodent that inhabits 3-state area in the U.S. RFLP for mtDNA of 87 individuals revealed 23 haplotypes Parsimony network reveals geographic relationships among haplotypes Haplotypes generally confined to single populations Major east-west split in distribution revealed Avise 2004 Problems with using Phylogenetics for Inferring Evolution It is a black box: starting from end point, reconstructing past based on assumed evolutionary model Orthologs versus paralogs Hybridization Differential evolutionary rates Assumes coalescence Gene Orthology Phylogenetics requires unambiguous identification of orthologous genes Paralogous genes are duplicated copies that do not share a common evolutionary history Difficult to determine orthology relationships paralogs paralogs Lowe, Harris, Ashton 2004 paralogs orthologs Gene Trees vs Species Trees Genes (or loci) evolve at different rates Why? Topology derived by a single gene may not match topology based on whole genome, or morphological traits Gene Tree B C A Coalescence Retrospective tracing of existing alleles to a common ancestral allele A reverse reconstruction of the evolution of modern variation Allows explicit simulation of sequence evolution Incorporation of factors that cause deviation from neutrality: selection, drift, and gene flow 9 generations in the history of a population of 14 gene copies Time present Slide courtesy of Yoav Gilad Individual alleles Gene Trees vs Species Trees Failure to coalesce within species lineages drives divergence of relationships between gene and species trees Divergent Gene Tree: Concordant Gene Tree b is closer to a than to c a b c b is closer to c than to a a b c How to model this process? Modeling from Theoretical Ancestors: Forward Evolution Can model populations in a forward direction, starting with theoretical past Fisher-Wright model of neutral evolution Very computationally intensive for large populations Alternative: Start at the end and work your way back Most recent common ancestor (MRCA) Time present Slide courtesy of Yoav Gilad Individual alleles The genealogy of a sample of 5 gene copies Most recent common ancestor (MRCA) Time present individuals Slide courtesy of Yoav Gilad The genealogy of a sample of 5 gene copies Most recent common ancestor (MRCA) Individual alleles Slide courtesy of Yoav Gilad Time present Examples of coalescent trees for a sample of 6 Time Individual alleles Slide courtesy of Yoav Gilad Coalescence Advantages Don’t have to model dead ends Only consider lineages that survive to modern day: computationally efficient Based on actual observations Can simulate different evolutionary scenarios to see what best fits the observed data Coalescent Tree Example Coalescence: Merging of two lineages in the Most Recent Common Ancestor (MRCA) Waiting Time: time to coalescence for two lineages Increases with each coalescent event E(Tk ) = 2N k(k -1) 2 Probability of Coalescence For any two lineages, function of population size Pcoalescence 1 2Ne Also a function of number of lineages Pcoalescence k (k 1) 1 2 2Ne where k is number of lineages Probability of Coalescence Probability declines over time Lineages decrease in number Can be estimated based on negative exponential Pcoalescence e k ( k 1) 1 t 2 2 Ne where k is number of lineages Average time to coalescence: 2N Time = k(k -1) 2 Time to Coalescence Affected by Population History Bottleneck Time to Coalescence Affected by Population History Population Growth How will population structure affect coalescence times? Time to Coalescence Affected by Population Structure Applications of the Coalescent Approach Framework for efficiently testing alternative models for evolution Inferences about effective population size Detection of population structure Signatures of selection Reconstructing history of populations Origins of Modern Humans Most fossil evidence points to origins in Africa and subsequent migrations Skulls found in Omo Valley, Ethiopia Dated at ~195K Omo 1 Modern http://wwwv1.amnh.org/exhibitions/ permanent/humanorigins /history/origin.php http://www.dhushara.com/book/unraveltree/unravel.htm Human Phylogeography: mtDNA Most ancient and diverse haplotypes in Africa (dots) Migration and admixture is evident from presence of African haplotypes in other clades Complexities to Human Phylogeography Sequence of X-linked ribonucleotide reductase M2 pseudogene 4 (RRM2P4) Coalescent evidence of Asian origin Strongly negative Tajima’s D in Europe due to low π How might you explain this? Garrigan 2007 Nature Reviews Genetics 7:669 Tajima’s D Expectations • D=0: Neutrality • D>0 d S – Balancing Selection: Divergence of alleles (π) increases OR – Bottleneck: S decreases • D<0 – Purifying or Positive Selection: Divergence of alleles decreases OR – Population expansion: Many low frequency alleles cause low average divergence