Page 204 - Applied Probability
P. 204

9. Descent Graph Methods
                              where j r = 0 or 1 according as the rth meiosis in j involves the transmission
                              of a grandmaternal gene or grandpaternal gene and similarly for k r .Ifwe
                              let θ i denote the recombination fraction between loci i and i + 1, then
                                                    m
                                       t i (k | j)  =    (1 − θ i ) 1 {k r −j r =0 mod 2} θ 1 {k r −j r =1 mod 2} .  189
                                                                          i
                                                   r=1
                               The meiosis indicators j and k can be viewed as elements of a commutative
                              group if we define addition by
                                         j + k  =[(j 1 + k 1 )mod 2,... , (j m + k m )mod2]
                              and the identity element by 0 =(0,..., 0). (See Problem 12 for the proper-
                              ties of a commutative group.) From the group perspective, Baum’s update
                              (9.13) becomes the convolution


                                         α i+1 (k)=     β i (j)t i (k − j)= β i ∗ t i (k),
                                                      j
                              where β i (j)= α i (j)φ i (y i | j). This suggests the possibility of exploiting
                              elementary Fourier analysis in the guise of the Walsh transform [16].
                                The Walsh transform turns a sequence a j indexed by a meiosis indicator
                              j into a new sequence ˆ a k indexed by a meiosis indicator k via
                                                        1      1   m

                                                    =                (−1) k r j r
                                                ˆ a k      ···              a j
                                                       j 1 =0  j m =0 r=1

                                                    =     (−1)  k,j  a j ,                (9.14)
                                                        j
                                             m

                              where  k, j  =     k r j r . The inverse Walsh transform
                                             r=1
                                                           1
                                                   ˇ a k  =      (−1)  k,j  a j .
                                                           2 m
                                                               j
                              deserves its names because
                                    1  	                  1
                                          (−1)  l,k  ˆ a k  =  (−1)  l,k   (−1)  k,j   a j
                                    2 m                  2 m
                                        k                    k          j
                                                         	     1
                                                     =      a j     (−1)  k,j−l
                                                              2 m
                                                          j       k
                                                                  1      1   m
                                                         	     1  	     	          k r (j r −l r )
                                                     =      a j      ···       (−1)
                                                              2 m
                                                          j      k 1 =0  k m =0 r=1
                                                     = a l .
   199   200   201   202   203   204   205   206   207   208   209