HC70A & SAS70A Winter 2009Genetic Engineering in Medicine, Agriculture, and LawTracking Human AncestryProfessor John NovembreThemes• Global patterns of human genetic diversity• Tracing our ancient ancestry• Clines versus clusters debate• Within-continent patterns • Personalized genomic ancestry inference • What really is ancestry?• Admixture and Chromosome painting• Natural selection and patterns of human genetic diversity • Salivary Amylase• Eye color (OCA2)Jin and Su, Nature Reviews Genetics (2000)Did archaic humans outside-of-Africa die out, only to be replaced later on or is this just an incomplete fossil record?Homo erectusA puzzle of human ancestryThe fossil record in East Asia contains a 60,000 year gapWhich of these is the best supported hypothesis for origins of humans around the globe?(A) Recent Out-of-Africa origin(B) Recent Out-of-Africa origin with minor contribution from ancestral archaic humans(C) Shared multi-regional evolution(D) Independent multiple originsCompeting models of human originsJin and Su, Nature Reviews Genetics (2000)Non-African sequences have a common ancestor at 25-75k years agoAll human mitochondrial DNA sequences have a common ancestor “Eve” 120-220k years agoMitochondrial “Eve” existed in the recent past!Tree shape is consistent with Out-of-Africa originHuman genome diversity panelA global-scale sample of human genetic diversityNot a perfect sampling, but the best studied yet...Review: Single nucleotide polymorphisms (SNPs) and haplotypesDNA Chips Can Detect SNP Genotypes (Or Haplotypes) Across An Individual’s Genome This Can Then Be Correlated With Diseases &/or Geographical AssociationsPopulation tree based on similarity of allele frequencies at ~650,000 SNPsSub-Saharan AfricaNorth Africa / Mid-EastEuropeCentral AsiaEast AsiaAmericasOceaniaLi et al, Science (2008)Population tree provides further support for “Out-of-Africa” origins of humansDistance via waypoints to: Addis Ababa, Rift Valley, Ethiopia(Putative origin of early modern humans) Decay of haplotype heterozygosity is consistent with a “serial bottlenecks” during Out-of-Africa expansionGlobal patterns of haplotype diversityExamples of paths from East Africa to each of the HGDP populationsInterpolated map of genetic diversity Decay of diversity is consistent with Out-of-Africa expansionMost Genetic Diversity Originated in the Founder Populations to Modern Humans!Summary: Human origins• At a global scale, genome-wide diversity patterns broadly consistent with a single, recent origin of modern humans in Africa• Further support available from recent Y-chromosome and autosomal gene TMRCA dates • Note: • A very small number of loci show very ancient TMRCA dates (e.g. 2 million years old). •Open question: Are these evidence for rare, ancient contributions from archaic humans?Continental-scale clusters of human variation?Results of an unsupervised clustering algorithm on microsatellite diversityGenetic “clusters” approximate continental-scale regionsRosenberg et al (2002) ScienceResults of unsupervised clustering algorithm on ~650,000 SNPs Clusters are now more detailedEurope, Middle East, and Central South Asia distinguishableContinental-scale clusters of human variation?Li et al (2008) ScienceOr does differentiation increase smoothly with geographic distance (“clines”)? Clines vs. clusters debatePerhaps clusters are due to impact of geographic barriers? SaharaTibetan PlateauBering StraitMalay archipelagoOr perhaps clusters are an artifact due to gaps in sampling?Either way: Variation is mostly found within rather than between groups Within populationsAmong population / within groupsAmong groupsLi et al (2008) ScienceSub-SaharanAfricaNorth AfricaMid-EastEuropeCentral AsiaEast AsiaOceaniaAmericasK = 2K = 3K = 4K = 5K = 6Admixture proportionsassuming K unknown populationsThe closer you look, the more “clusters” you see!Jakobsson et al (2008) NatureImage from EU Panorama “27 Europeans”Fine-scale analysis within a continent: Europe as a case study1400 individuals from 37 unique populations in EuropeAnalysis using 197,000 SNP lociSingle Nucleotide PolymorphismsIndividualsPC1PC2-0.10-0.050.0 00.0 50.1 00.1 5-0.06 -0.04 -0.02 0.0 0 0.0 2 0.0 4 0.0 6Pop. 1 Pop. 2Admixed IndividualsPrincipal Components Analsysis (PCA)=Reduction of dimensions......0 = AA1 = AB2 = BBPC2-0.080.10-0.050.050.08-0.020.04-0.03-0.050.070.08-0.02-0.040.06-0.020.050.08-0.02-0.040.03...PC10.0550.042-0.0520.0100.0520.043-0.0580.020-0.030-0.0500.0520.043-0.058-0.040.020-0.031-0.0530.0140.533-0.044...PCA is often a useful tool to visualize patterns of population