edwards

MBI 2005 Workshop in Phylogenetics and Phylogeography: 

MBI 2005 Workshop in Phylogenetics and Phylogeography Perspective an empiricist (and bird watcher!) Enthusiastic producer of DNA sequence data unabashed consumer of computational analysis packages Allele frequency approaches to population structure Analysis of population structure in House Finches using AFLPs Diagnosing populations with large numbers of loci Geographic differentiation: contrast between statistical and biological significance Genealogical approaches to population history Multilocus sequence data sets for Australian birds Little reciprocal monophyly for neutral loci Need for multi-species population genetic models

House Finches and Mycoplasma:a recent host-parasite interaction: 

House Finches and Mycoplasma: a recent host-parasite interaction Mycoplasma gallisepticum escaped chickens and invaded House Finches in the eastern U. S., ~1994 9 years later, finches are more resistant to the bacterium and recent strains are attenuated Have finches evolved resistance? AFLP study: What is the geographic setting in which this epizootic took place?

Recent history of House Finch populations: 

Recent history of House Finch populations ~1870 bottleneck? 1940 ~200 birds historic range

AFLP survey of House Finches: 

AFLP survey of House Finches California Mexico Purple Finch Cassin’s Finch 163 individuals, 16 populations 3 primer combinations 166 polymorphic bands 61% polymorphic bands Wang Z, et al. (2003) Evolution 57, 2852-2864.

Slide5: 

0 1 2 3 4 5 6 7 8 9 10 CA CA AR TX CO WA HI Can. NY MI ME OH MD PA AL 70.7% 8.1% Nucleotide diversity (estimated number of substitutions per 1000 sites) Distribution of variation (AMOVA) Among individuals w/in pops. Among pops. w/in subspecies Mex. original range Among subspecies 21.2% introduced range AFLPs: House Finch are moderately structured with little evidence for genetic bottlenecks Wang Z, et al. (2003) Evolution 57, 2852-2864.

Slide6: 

Western U.S. Hawaii Eastern U. S. Tripartite structure of House Finch populations suggested by assignment test of AFLP data (program STRUCTURE: J. Pritchard et al. 2000. Genetics 155: 945-959) Wang Z, et al. (2003) Evolution 57, 2852-2864.

Fst with large numbers of loci: statistical and biological significance: 

Fst with large numbers of loci: statistical and biological significance Wang Z, et al. (2003) Evolution 57, 2852-2864. Geogaphic distance (km)

Slide8: 

Low resolution of population trees using AFLP data Wang Z, et al. (2003) Evolution 57, 2852-2864.

1. Gene trees don’t always match the species tree2. Gene divergence often precedes population divergence: 

1. Gene trees don’t always match the species tree 2. Gene divergence often precedes population divergence Gene tree Species tree T/2Ne 2Ne Two rules of gene trees near the species boundary

Slide10: 

present Time [generations] Time [generations x m = number of mutations] past t T  /2 D  The need for estimating ancestral population size () when inferring species divergence time ( or )

Slide11: 

1 2 3 1 3 2 2 3 1 Nei 1987. Molecular Evolutionary Genetics Wu 1991. Genetics 127:429-435 Hudson 1992. Genetics 131:509-512 Yang 2002. Genetics 162:1811-1823 Rannala B, Yang Z. 2003. Genetics 164, 1645-1656. congruent gene tree incongruent gene tree incongruent gene tree Species vs. Gene Trees

Slide12: 

- multiple independent loci estimates ancestral q (present q also) - estimates population divergence times - uses branch length information - accounts for uncertainty in gene trees Assumptions: “know” the species tree - random mating - no gene flow after population divergence - free recombination among loci (not within) Bayes Markov chain Monte Carlo (MCMC) method (Yang and Rannala) Yang 2002. Genetics 162:1811-1823 Rannala B, Yang Z. 2003. Genetics 164, 1645-1656

Slide13: 

pUC 18 vector 1-2 kb insert Genomic DNA Obtaining anonymous loci Mutational scale directly comparable to mtDNA Mutational homoplasy is minimal Gene trees easily constructed Diversities easily compared across species Advantages of SNPs over microsatellites Brumfield R. et al. 2003TREE 18, 249-256.

Slide14: 

Diversity Locus 1 2 3 4 5 6 7 8 * * * * Locus chosen for phylogeographic survey Microsatellites SNPs Ascertainment bias and sampling strategy for SNPs Diversity Locus 1 2 3 4 5 6 7 8 * * * * * * * panel study population SNP

Life in the Outback: 

Life in the Outback

Slide16: 

field sampling Demographic inference Pipeline for multilocus data analysis

Resolving haplotypes from diploid PCR products: 

Resolving haplotypes from diploid PCR products AGGCGT CTTACGACCTTAAGCTCATC GATAATCTC A T G C diplotype

Slide18: 

Resolving heterozygous indels via cloning or allele-specific PCR Jennings WB, Edwards SV (2005) Evolution in press.

Concordance of geographic ranges of Australian songbirds: 

Concordance of geographic ranges of Australian songbirds

Carpentarian barrier (B) is deepest split in area cladograms: 

Carpentarian barrier (B) is deepest split in area cladograms Cracraft J (1986) Evolution 40, 977-996.

Gene tree of western babbler lineage: 

Gene tree of western babbler lineage Migration event reconstructed by parsimony

Assumed phylogeny of Poephila finches: 

Assumed phylogeny of Poephila finches hecki acuticauda c. atropygialis Long-tailed Finch Black-throated Finch

Characteristics of 30 anonymousloci from Poephila finches: 

Characteristics of 30 anonymous loci from Poephila finches 0 1 2 3 4 5 6 7 8 9 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800 More Sequence length (bp) Frequency 0 1 2 3 4 5 6 7 0 0.4 0.8 1.2 1.6 2 2.4 2.8 3.2 3.6 4 More Percent divergence Frequency 0 1 2 3 4 5 6 7 30 33 36 39 42 45 48 51 54 More % GC content Frequency Jennings WB, Edwards SV (2005) Evolution in press.

30 gene trees from Australian finches: 

30 gene trees from Australian finches Jennings WB, Edwards SV (2005) Evolution in press.

Four parameters estimated: 

Four parameters estimated

Slide26: 

prior, analysis 1 prior, analysis 2 posterior, analysis 1 posterior, analysis 2 ML estimate

Decreasing variance with increasing numbers of loci - mostly: 

Decreasing variance with increasing numbers of loci - mostly

Slide28: 

Analysis suggests Pleistocene divergence across the Carpentarian barrier  = 4Ne  = t generation time = 1 year = ~3.6 x 10-9 substitutions/site/year (gamebirds) hac Ne = 521,000 (320,000-767,000) hac(t) 0.1 MY (0.01 MY - 0.34 MY) hecki acuticauda c. atropygialis “Long-tailed Finch” “Black-throated Finch” ha Ne = ~384,000 (240,000-530,000) ha (t) 0.61 MY (0.35 MY - 86 MY)

Slide29: 

0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60 1.80 2.00 2.20 2.40 2.60 2.80 3.00 1 Divergence Time (MYA) Pleistocene Pliocene gene divergence ( D / 2) population divergence ( g ) - MLE population divergence ( g ) - Bayesian cincta vs. (acuticauda, hecki) acuticauda vs . hecki Gene divergence (D/2) Population divergence () - MLE Population divergence () - Bayesian

Slide30: 

hecki Kimberley/ Arnhem Land Barrier (0.6 MYA) maps from Schodde and Mason 1999. The Directory of Australian Birds: Passerines Black-throated Finch Long-tailed Finch Summary of Population Divergence Times Carpentarian Barrier 0.7 MYA acuticauda atropygialis

East-west pairs -- treecreepers: 

East-west pairs -- treecreepers

Slide32: 

AL14 + + + + AL16 AL18 AL19 AL20 + + AL21 AL22 + + + + AL3 AL5 + AL7 Brown Treecreeper (eastern) lineages Black-tailed (western) Treecreeper lineages 10 gene trees in Australian treecreepers Pilbara (far west) Cape York (far north)

Migrate analysis reveals extensive gene flow and geographic variation in population size: 

Migrate analysis reveals extensive gene flow and geographic variation in population size Newman Longreach Bourke Forsayth Sedan Douglas Hot Springs Fitzroy Crossing Black-tailed Treecreepers Brown Treecreepers Doomadgee 0.0047 +/-0.0014 Weipa Rotzel, Edwards and Beerli, unpubl. data = 0.0102

Implications of finch gene trees for selection and thegenealogical species concept: 

Implications of finch gene trees for selection and the genealogical species concept Edwards et al. 2005 PNAS 102, 6550 acuticauda, hecki cincta

Alternative models of population history: quo vadis?: 

Alternative models of population history: quo vadis? equilibrium migration model pure isolation model isolation-migration model

Slide36: 

Conclusions • As the number of loci grows, distinguishing statistical from biological significance will be important • Lack of reciprocal monophyly will be common in vertebrate sequence data sets • Is haplotype inference really necessary?

Acknowledgements: 

Acknowledgements Funding NSF Australian phylogeography Bryan Jennings Nancy Rotzel Peter Beerli House Finch evolution Geoff Hill Zhenshan Wang