BSCI222 – Lecture 25 (December 12, 2013)- Chapter 26: Evolutionary Genetics. Neutral theory, molecular clocks, and the history of time.- Techniques for looking at protein polymorphisms in the 60s revealed astonishing levels of genetic variation. Until then, the standard view of genotype was wild-type for most loci, and mutant for one/a few (called wild-type at most loci view). When actually measured the genetic variation, turns out that everybody is heterozygous at almost every locus.o Starch gel electrophoresis: heat up potato starch with a buffer until boiling  mold  create gel  cut and insert filter paper pieces dipped in something that has been sonicated in order to release all the DNA, each piece of paper from a separate individual. Do the electrophoresis  slice the gel horizontally so that you can make many replicate slices, and need to stain the gel in order to see the proteins and in order to visualize individual loci from the huge smear. Layer manydifferent kinds of stains linked to chromogenic precipitates. Each “lane” contains the proteins from a different individual, and the stain is detecting a particular enzyme, particular biochemical activity. Different bands are different allelic forms of the protein, causing it to migrate at different rates through the gel because of different charge. One allele per row. Number of rows = number of alleles. Each column is one person; one dot is really AA or CC, depending on what row it’s in, even though it’s just one dot. Don’t know which is the wild-type allele, because all at a high frequency.o Demanded that the theory be revised.  First stage: denial. Tried explaining these results as an oddity, from a balancing selection (selection that prefers heterozygotes). Not true, because found no matter what loci you look at. A model of selection: - 3 genotypes: AA (frequency of p^2, fitness of 1) Aa (frequency of 2pq, fitness of 1-s) and aa (frequency of q^2, fitness of 1-2s)- Hardy-Weinberg: the fraction of selective deaths each generation  D = 2pq(s) + q^2(2s). If s = 0.1, and p and q are both 0.5, then 10% of the population is removed by selection in one generation. This selection changes the allele frequency, slightly. Have to go through many generations of this selection in order to achieve the cumulative number of deaths necessary to fix an allele (fixation, D = -2logp).- If you have a new mutant, and it starts at 10^-6 frequency, then D = 27.6 (times the population size). Motoo Kimura: wanted to demonstrate that polymorphisms couldn’t be fixed in the population by selection. Found a value so high that he concluded it couldn’t possibly be due to selection. The rate of proteinevolution depends on WHAT protein; some are highly conserved and change in their sequence (one amino acid change per 20 million years in cytochrome c, very important for metabolism) rarely over the phylogenetictree. Fibrinopeptides, the sequence doesn’t matter much, evolve very rapidly. Kimura came up with an average rate of amino acid substitution of 1% per 28 million years. How many nucleotide substitutions do we have to have over the whole genome, to explain these rates of protein evolution?- Assumed haploid genome size of 4 x 10^9 bp, triplet genetic code, and 20% silent substitutions. Then the rate of nucleotide substitution over the whole genome = 0.57 substitution/per year (or1 substitution/2 years over the whole genome). - Plug this into Haldane’s formula: 27.6 populations for every substitutions  kill of 15.8 times the population every year. That’sa problem, obviously! Conclusion: most substitutions must be selectively neutral, and most mutations are fixed by random genetic drift and not by selection. Shocking and scandalous at the time.- Null hypothesis for molecular evolution: Neutral Mutation – Random Drift Hypothesis: E (evolution) = M (rate of mutation)*F (rate of fixation). This is how geneticisits define evolution, as a change in allele frequency. Rate of mutation is 2N (mu). Probability of fixation = 1/2N (more likely in smaller population). E = 2N (mu) * 1/2N = mu. Rate of evolution = rate of mutation. Independent of the population size!- What is neutral? Neutral depends on population size. In a small population, even a selectively favored allele can be lost due to random drift. Takes a big selective coefficient to cause something to be fixed in a small population. If the selection coefficient (s) is less than or equal to 1/N (population size), then will be neutral and these weakly selected alleles are mostly affected by drift. - The Molecular Clock: substitutions should accumulate steadily if the rate of evolution is determined primarily by the rate of mutation (exponentially). o The observed rate of evolution is only slightly lower than the observed rate of mutation.o This has become a tool for annotating genome sequences. If you can compare genomes and look at the amount of divergence in little windows across the genome, can identify regions that are evolving very rapidly or very slowly. The slowly ones are the functional bits that we might be interested in, the coding sequences. The highest rates of substitution, with very little selection, are the silent/synonymous areas (3rd amino acid, etc.) and pseudogenes (no longer functional, copies). Very slowest rates are in the nonsynonymous substitutions inthe coding regions of genes; highly examined by selection and made sure it doesn’t happen. Different parts of genes evolve at different rates, depending on the degree of selective constraint.  But, the rate of evolution is pretty much the same across the genome for the synonymous sites. Histones essentially have a rate of 0.0, even over billions of years of evolution.o Very useful tool for reconstructing the history of life; have fossils in the genome. Interested in: history of relationship among species, and the history of the genes and gene families inside our genome. Early favorite molecule for doing this was mitochondrial DNA. Was easy to isolate, didn’t have to clone because of the high copy number and beingcircular make it easy to separate. Also, because it’s passed from mother to offspring without recombination, then you get a perfect clonal inheritance (not exchanging unrelated bits through recombination). Just gets passed down, and occasionally experiences substitutions over the years. Essentially making a phylogenetic tree; number of differences tells you