BIOL 2007- POPULATION GENETICS TUTORIAL - Single locus selection and drift
Try out these problems from 2001!  Then look at the answers.

Use lecture handouts when necessary. Ring tutor or J. Mallet when all else fails!

1) In a study of evening primroses, Levy and Levin (1975) used electrophoresis to study the gene for phosphoglucose isomerase-2 in different strains.  They found the following genotypes:

`A1A1         A1A2         A2A2         Total`
` 35          19            3           57`
a) Estimate the allele frequencies at this locus, and then, using the Hardy-Weinberg law, find the expected genotypic frequencies.

b) Use these frequencies to estimate the expected numbers of genotypes, and test for deviation from Hardy-Weinberg using a chi-square goodness of fit test [Note:  , where O=observed numbers, E=expected numbers]; what is the total chi-square value?

c) Look up the probability of getting results this extreme in the following simplified chi-square table (Note: for a worked example and problems with degrees of freedom, see the lecture on "Evolution of Genetic Diversity"). Write down this P-value.

______________________________________________________________
`Degrees of           Value of P`
`freedom              0.99  0.9   0.5   0.1   0.05  0.01  0.001`
`1                    0.00  0.02  0.46  2.71  3.84  6.63  10.83`
`2                    0.02  0.21  1.39  4.61  5.99  9.21  13.82`
d) Is there significant evidence for deviation from Hardy-Weinberg? Write down three possible causes for a deviation (whether there is or isn't).

2) Phenylketonuria is an autosomal recessive form of severe mental retardation. About 1 in 10,000 newborns are affected.

a) Assuming random mating, what is the expected frequency of heterozygous carriers assuming equilibrium between mutation and selection has been reached?
If phenylketonuria is at the equilibrium of mutation/selection balance, we expect the equilibrium frequency q* to be  ( is the mutation rate, s is the selection coefficient; see notes on mutation/selection balance, or follow the link on this subject from "Evolution of Genetic Diversity" lecture, for an explanation).

b) Assuming a typical mutation rate of , what is the selection pressure, s, against suffererers?  What is the fitness?  Does this value seem sensible?

3) Inbreeding causes a deficit of heterozygotes in the population. Supposing the deviation from Hardy-Weinberg in question 1 is caused by inbreeding. What is the value for F? HINT:Observed heterozygote frequency = 2pq (1-F). Rearrange to solve for F.

4) It has been suggested that we should not allow people who have deleterious phenotypes to reproduce. Suppose a gene causing a recessive disease has a frequency of 0.05 in the population.

a) Assuming random mating, what is the fraction showing the disease?

b) Using the formula given in the "selection and the single gene" lecture, what will be the change in gene frequency in a single generation (Dp) if we prevent such people reproducing? (This is equivalent under natural selection to killing off the recessive phenotypes, so that s=1. (HINT:the change in frequency of the recessive allele has the same magnitude but is opposite in sign to the change in the dominant allele frequency, if you think about it!).

c) What is the new gene frequency after one generation? What will be the new genotype frequency assuming Hardy Weinberg?

d) Is the approximation given in the lecture notes good for so large a frequencyas p=0.05? (See the lecture notes, "Selection and the single gene", under "How fast do populations respond?")

e) Explain whether you think it is a good idea to prevent people with recessive genetic diseases from having children.

5) My Dad is my Mum's first-cousin-once-removed (see diagram on left; I am marked with a *).

a) What is my inbreeding coefficient, F? (i.e. the probability that I have a diploid allele at any locus that is identical by descent through my grandparents. Ignore sex-linkage).

b) Is it a bad thing to be so inbred?