2006 by The University of Chicago. All rights reserved.
Appendix from A. F. Agrawal and S. P. Otto, "Host-Parasite
Coevolution and Selection on Sex through the Effects of Segregation"
(Am. Nat., vol. 168, no. 5, p. 617)
Results from Generalized Models of Infection
Here we generalize the MA, IMA, and GFG models of infection to allow more flexibility in describing the
susceptibility of heterozygous hosts and the infectiousness of heterozygous parasites. To maintain generality, we
use four parameters describing the gene expression of heterozygous hosts: dHA is the probability that an Aa host
expresses just the A allele (and has the same pattern of susceptibility as an AA host); dHAa is the probability that
an Aa host expresses both alleles; dHa is the probability that an Aa host expresses just the a allele (and has the
same pattern of susceptibility as an aa host); and dH is the probability that an Aa host expresses neither allele.
Note that . Similarly, we use the parameters dPA, dPAa, dPa, and dP to describe thed d d d p 1HA HAa Ha H
analogous properties for heterozygous parasites.
We also generalize the GFG model to allow heterozygotes to bear different costs. Specifically, we set the cost
of resistance in heterozygous hosts to , where hcc is the cost of expressing both alleles together,(d d h )cHA HAa c
which might not be as great as the cost, c, of expressing only the resistant allele, A. Similarly, we set the cost of
infectiousness in heterozygous parasites to , where hkk is the cost of expressing both alleles(d d h )kPa PAa k
together, which again might not be as great as the cost, k, of expressing only the infectious allele, a.
By using these parameters in conjunction with the logic underlying each model of infection, we can generate