1	Allele Frequencies & Allele Frequency Change Change of allele abundance/frequency is the basic evolutionary change.
2	Reading Assignment Chapter 24 Evolutionary Processes is the reading assigned for this lecture. Frequency is used in the sense of a proportion, frequency takes values between 0 and 1.
3	Counts and Frequencies The increase in abundance is a true measure of fitness. Most of population models deal with frequency, ranging from 0 to 1, of two or more types, i.e., relative abundance. We will continue in that tradition.
4	Genotype frequencies are determined by counting the number of each type and dividing by the total.
5	Genotype frequencies
6	Phenotype frequencies If the A locus has classic dominance, there are only two phenotypes, which we can symbolize as A_ and aa. The frequency of the A_ (the dominant phenotype) is 0.24 + 0.41 = 0.65; the frequency of the homozygous recessive phenotype is the same as the homozygous recessive genotype = 0.35
7	Allele (or Gene) Frequencies
8	Instead of 17 individuals think of the population as 34 (2x17) genes
9	RANDOM MATING Random mating means that mating pairs are drawn independently from the available pool of individuals. Independent means Pr(AA x aa mating) = Pr(AA)•Pr(aa), etc. With some algebra I will not show, it can be shown that random mating is equal to a random pick of alleles from a gene pool.
10	RANDOM MATING
11	IF Mating is Random, then Expected genotype frequencies can be predicted from allele frequencies. e.g. freq(I^B i) = 2•freq(I^B)•freq(i) If genotype frequencies can be predicted then phenotype frequencies are also predicted. e.g. freq(B blood type) = 2•freq(I^B)•freq(i) + freq(I^B)• freq(I^B)
12	A worked example using ABO Given freq(I^A) = 0.3, freq(I^B) = 0.1 so freq(i) = 1 -0.3 -0.1 = 0.6 (as they must add to 1) THEN the expected frequency of the I^Bi genotype (using Hardy-Weinberg) is 2•.1•.6 = 0.12 The expected frequency of the B phenotype is expected freq(I^Bi) + freq(I^BI^B) = 2•.1•.6 + .1•.1 = 0.13 or 13%
13	Expected genotype frequencies Knowing the allele frequencies in the current generation and that mating is random with respect to the genotype of interest allows one to calculate the expected genotype frequencies in the next generation (Hardy-Weinberg). The observed frequencies will rarely equal the expected frequencies.
14	Differences between expected and observed are evaluated with Chi-squared test If you are told that p = 0.6 (and therefore q must equal 0.4) and you observe the following numbers of AA, Aa & aa, respectively 30, 40, 10 You note that 80 individuals were observed and calculate the expected numbers as N•p², … = 28.8, 38.4, 12.8
15	Chi-squared calculation Χ²= 1.2²/28.8 + 1.6²/38.4 + 2.8²/12.8 Χ²= 0.05 + 0.067 + 0.6125 Χ²= 0.73 Χ² is less than the critical value, 5.99, with 2 degrees of freedom so hypothesis is accepted. The degrees of freedom (df) in this problem are 2, because you were given the gene frequencies and the observed numbers. Normally one has to estimate allele frequency from data. If that is done there is one less df.
16	Estimating allele (or gene) frequencies from phenotypic frequencies in cases with dominance.
17	Allele frequencies can be estimated from phenotype frequencies (but only if one assumes mating is random).
18	Species have geographic structure Species are typically divided into many populations which are clumped in space. Allele frequencies are likely to be different among populations of a species. Movement of individuals between populations is known as gene flow (also called migration.)
19	Movement among populations
20	Directional Selection of Alleles If an allele increases in frequency no matter what its initial frequency, it is considered to be directionally selected. The most clear examples of directional selection are examples of DDT resistance and antibiotic resistance.
21	Phenotype Selection Patterns Directional selection at phenotypic level results when survival increases (or decreases) in increasing phenotypic value. Stabilizing selection is the pattern where the middle phenotypic values have the greatest survival and the extremes have the least.
22	BIOLOGICAL POPULATIONS ARE FINITE The calculation of expected genotype frequencies given random mating assumes that p does not change with time. That is not possible with finite populations. All real populations have a finite number of members, and even if mating is independent of genotype, the actual numbers change slightly from the expected due to sampling and the population size, N, becomes an important variable.
23	Vocabulary Allele frequency Genotype frequency Phenotype frequency Random mating Expected genotype frequency Hardy-Weinberg Directional selection Stabilizing selection Finite