Transcript ppt
Lecture 8: Selection in Real Populations September 18, 2015 Announcements u Minor schedule changes Added a selection lecture (today) Delayed drift lecture (next Friday) Merged population structure lectures u Exam 1 is Wednesday, Sept 23 in computer lab. u Review session on Monday: bring questions u Sample exam and key are posted on website Last Time Dominance and types of selection Why do lethal recessives stick around? Equilibrium under selection Stable equilibrium: overdominance Unstable equilibrium: underdominance Today Underdominance and Overdominance revisited Overview of advanced topics in selection Introduction to Genetic Drift Why does “nontrivial” equilibrium occur with underdominance? Why doesn’t A1 allele always go to fixation if A1A1 is most fit genotype? ω Proportion of A1 alleles in heterozygous state: pq (pq+p2) = q A1A1 A1A2 A2A2 What determines the equilibrium point with underdominance? s1=0.2; s2=0.2 ω11=1; ω12=0.8; ω22=1 ω s1=0.2; s2=0.05 ω11=0.85; ω12=0.8; ω22=1 Allele Frequency (q) Why does equilibrium point of A1 allele frequency increase when selection coefficient decreases? A 1A 1 A1A2 A2A2 s1 qeq = s1 + s2 s1 peq s2 qeq Example: Kuru in Fore Tribespeople Prion disease in Fore tribesmen Transmitted by cannibalism of relatives by women/children Cannibalism stopped in 1950’s Older people exposed to selection, younger are ‘controls’ Identified locus that causes susceptibility: Prion Protein Gene, PRNP MM and VV are susceptible, MV are resistant http://learn.genetics.utah.edu/features/prions/kuru.cfm Genotype Counts at Postion 129 of the PRNP Gene Selection MM MV VV No Cannibalism 31 30 37 Cannibalism 4 23 3 Are populations in Hardy-Weinberg Equilibrium without selection? How about with selection? What type of selection appears to be occurring? Effects of Kuru on Viability of Different Genotypes For overdominance, relative viability can be estimated as the proportion of a genotype that survives to adulthood, relative to survival of the heterozygote (see Hedrick p. 144 for derivation) PP n11 = PP ' 11 12 ' 11 12 P P n 22 = P P ' 22 12 ' 22 12 Kuru and Heterozygote Advantage 1 v (s ) Selection coefficient 2 (only females) 0.403 0.2985 0.373 P P (0.133)(0.514) n11 = = = 0.403 incorrect in text! PP (0.221)(0.767) ' 11 12 ' 11 12 sMM qeq 0.483 sMM sVV Balancing selection maintains polymorphism in human populations Directional selection predominates for most loci Why doesn’t selection quickly wipe out most variation? Antagonistic Pleiotropy Individual alleles affect multiple traits with opposing effects on fitness components Aspen and elk herbivory in Rocky Mountain National Park Aspen can inhibit herbivory with protective compounds: phenolic glycosides Tradeoff with growth Osier and Lindroth, Oecologia, in press Phenolic glycosides (%) How does selection work in a variable environment? Spatial versus temporal variation Spatial variation maintains diversity, especially if habitat choice occurs Temporal variation less effective at maintaining diversity, except for perennials Example: plants that rely upon flooding for establishment often have large variation in flowering phenology. Early or late flowering can be favorable depending on timing of snow melt and spring floods. Industrial Melanism Peppered moth (Biston betularia) has dominant dark morph Elevated frequency in polluted areas Frequency of dominant morph declining with environmental cleanup Rate of decline modeled with basic selection model, s=0.153 http://www.leps.it/in dexjs.htm?Speci esPages/BistoBet ul.htm Frequency Dependent Selection Relative fitness is a function of frequency in the population Negative frequency-dependence: fitness is negatively correlated with frequency Should maintain variation in the population Examples include predator-prey interactions, pollinatorfloral interactions, and differential use of nutrients by different genotypes Positive frequency-dependence: fitness is positively correlated with frequency Should drive alleles to fixation/loss more rapidly Examples include decreased pollination for rare flowers, or increased predation for unusual phenotypes Frequency Dependent Selection in an Orchid Dactylorhiza sambucina has yellow and purple morphs No nectar or pollen reward for pollinators Naive pollinators switch to different flower color if no reward provided Rare color morphs favored http://www.treknature.com/gallery/Euro pe/Czech_Republic/photo9844.htm Frequency Dependent Selection in a Fish Perissodus microlepis is scale-eating cichlid fish from Lake Tanganyika in central Africa Assymetrical jaw causes feeding on alternate sides of prey Frequency of left-and right jawed morphs fluctuates around 0.5 Prey are on lookout for more common morph http://bio.research.ucsc.edu/~barrylab/ classes/evolution/Image61.gif Coevolution Organisms exert selection pressure on each other, evolve in response to each other Pest and pathogen Predator and prey Competitors Mutualists Maintains variation in both species through time Red Queen Hypothesis http://en.wikipedia.org Coevolution and the Importance of Sex Evolution of sex is mystifying: sacrifices 50% of fitness and breaks up adaptive genotypes Red Queen hypothesis is one explanation: stay ahead of the pathogens by generating variation Experiment in C. elegans: Infect populations with pathogen Serratia marcescens Selfing populations have higher mortality long-term Morran et al. 2011 Science 333: 216-218 How will the frequency of a recessive lethal allele change through time in an infinite population? What will be the equilibrium allele frequency? What Controls Genetic Diversity Within Populations? 4 major evolutionary forces Mutation Drift + - Diversity +/- Selection + Migration Genetic Drift Relaxing another assumption: infinite populations Genetic drift is a consequence of having small populations Definition: chance changes in allele frequency that result from the sampling of gametes from generation to generation in a finite population Assume (for now) Hardy-Weinberg conditions Random mating No selection, mutation, or gene flow Genetic Drift A sampling problem: some alleles lost by random chance due to sampling "error" during reproduction