Page 34 - Applied Probability
P. 34

17
                                                       1. Basic Principles of Population Genetics
                                 9. To verify convergence to linkage equilibrium for a pair of X-linked
                                   loci A and B, define P nx (A i B j ) and P ny (A i B j ) to be the frequencies
                                   of the A i B j haplotype at generation n in females and males, respec-
                                   tively. For the sake of simplicity, assume that both loci are in Hardy-
                                   Weinberg equilibrium and that the alleles A i and B j have frequencies
                                   p i and q j .If z n denotes the column vector [P nx (A i B j ),P ny (A i B j )] t
                                   and θ the female recombination fraction between the two loci, then
                                   demonstrate the recurrence relation
                                                                          1

                                                                          2
                                                    z n  = Mz n−1 + θp i q j               (1.5)
                                                                         1
                                   under the usual equilibrium conditions, where the matrix
                                                                1 [1 − θ]  1

                                                      M   =     2        2  .
                                                                 1 − θ   0
                                   Show that equation (1.5) can be reformulated as w n = Mw n−1 for
                                                  t
                                                                   t
                                   w n = z n − p i q j 1 , where 1 =(1, 1) . Solve this last recurrence and
                                   show that lim n→∞ w n = 0. (Hints: The matrix power M n  can be
                                   simplified by diagonalizing M. Show that the eigenvalues ω 1 and ω 2
                                   of M are distinct and less than 1 in absolute value.)
                                10. Consider an autosomal dominant disease in a stationary population.
                                   If the fitness of normal A 2 /A 2 people to the fitness of affected A 1 /A 2
                                   people is in the ratio 1 − s : 1, then show that the average num-
                                   ber of people ultimately affected by a new mutation is  1−s  . (Hints:
                                                                                     −s
                                   An A 2 /A 2 person has on average 2 children while an A 1 /A 2 person
                                   has on average  2  children, half of whom are affected. Write and
                                                  1−s
                                   solve an equation counting the new mutant and the expected num-
                                   ber of affecteds originating from each of his or her mutant children.
                                   Remember that s< 0.)
                                11. Consider a model for the mutation-selection balance at an X-linked
                                   locus. Let normal females and males have fitness 1, carrier females
                                   fitness t x , and affected males fitness t y . Also, let the mutation rate
                                   from the normal allele A 2 to the disease allele A 1 be µ in both sexes.
                                   It is possible to write and solve two equations for the equilibrium
                                   frequencies p ∞x and p ∞y of carrier females and affected males.

                                    (a) Derive the two approximate equations
                                                                       1
                                                     p ∞x  ≈ 2µ + p ∞x t x + p ∞y t y
                                                                       2
                                                                      1
                                                     p ∞y  ≈ µ + p ∞x t x
                                                                      2
                                        assuming the disease is rare.
   29   30   31   32   33   34   35   36   37   38   39