Page 97 - Applied Probability
P. 97

5
                              Genetic Identity Coefficients
                              5.1 Introduction

                              Genetic identity coefficients are powerful theoretical tools for genetic analy-
                              sis. Geneticists have devised these indices to measure the degree of inbreed-
                              ing of a single individual and the degree of relatedness of a pair of relatives.
                              Since the degree of inbreeding of a single individual can be summarized by
                              the relationship between his or her parents, we will focus on identity coef-
                              ficients for relative pairs. These coefficients pertain to a generic autosomal
                              locus and depend only on the relevant pedigree connecting two relatives
                              and not on any phenotypes observed in the pedigree. In Chapter 6 we will
                              investigate the applications of identity coefficients. Readers desiring moti-
                              vation for the combinatorial problems attacked here may want to glance at
                              Chapter 6 first.



                              5.2 Kinship and Inbreeding Coefficients

                              Two genes G 1 and G 2 are identical by descent (i.b.d.) if one is a physical
                              copy of the other or if they are both physical copies of the same ancestral
                              gene. Two genes are identical by state if they represent the same allele.
                              Identity by descent implies identity by state, but not conversely. The sim-
                              plest measure of relationship between two relatives i and j is their kinship
                              coefficient Φ ij . Mal´ecot [12] defined this index to be the probability that a
                              gene selected randomly from i and a gene selected randomly from the same
                              autosomal locus of j are i.b.d. The kinship coefficient takes into account
                              the common ancestry of i and j but not their observed phenotypes at any
                              particular locus. When i and j are the same person, the same gene can
                              be drawn twice because kinship sampling is done with replacement. The
                              inbreeding coefficient f i of an individual i is the probability that his
                              or her two genes at any autosomal locus are i.b.d.; inbreeding sampling is
                                                                 1
                              done without replacement. Since Φ ii = (1+f i ) and f i =Φ kl , where k and
                                                                 2
                              l are the parents of i, an inbreeding coefficient entails no new information.
                              Note that f i = 0 unless i’s parents k and l are related. If f i > 0, then i is
                              said to be inbred.
                                The last column of Table 5.1 lists kinship coefficients for several com-
                              mon types of relative pairs. The table also contains probabilities for other
                              identity coefficients. Before defining these additional indices of relationship,
                              let us focus on a simple algorithm for computing kinship coefficients. This
   92   93   94   95   96   97   98   99   100   101   102