Pedigreed crosses to estimate recessive virulence allele frequencies in natural populations of gall midges

Monitoring changes in rare, recessive allele frequencies in natural populations can be accomplished using pedigreed individuals sampled from these populations. A pedigree keeps track of and limits the mating of sampled individuals, to preserve information about the genotype of the sampled individual in the phenotypes of its descendents. To estimate allele frequencies in a natural population using pedigreed crosses, four relations must be specified: (1) a method to determine whether the pedigreed line carries the desired allele; (2) a method to estimate the phenotypic frequency of the trait among the pedigreed lines and a credibility limit for the estimate; (3) the genetic relation between the phenotype frequency among the lines and the allele frequency in the natural population; and (4) a method to estimate the probability that the first method did not detect the trait, assuming that the allele was present in the sampled individual. Knowledge about the segregation patterns of the allele enables specification of (3) and (4). Bayesian statistics were used to estimate the phenotypic frequency of the trait among the pedigreed lines. The method determining whether the pedigreed line carries the desired allele will vary with the species and trait of concern. We focused on monitoring of vGm1, a recessive autosomal allele, and vGm2, a recessive sex-linked allele, which provide virulence against certain rice resistance genes in rice gall midge, Orseolia oryzae (Wood-Mason) (Diptera: Cecidomyiidae). We show how three pedigrees can be used to estimate these allele frequencies. An F(1) field screen challenges the F(1) offspring of sampled individuals on the rice differentials. A P(1) test-cross mates the sampled individual with a homozygous lab colony for the allele of interest, and evaluates their offspring on the rice differentials. A conditional F(1) test-cross takes the offspring from pedigrees that were negative in an F(1) field screen, and test-crosses these offspring with the homozygous laboratory colony. We also indicate how to test for independent assortment when a double (or multiple) homozygote laboratory colony is used in a test-cross, how to test for differences among samples, and how to pool data to produce a single estimate based on a larger number of pedigreed lines. These methods may encourage the development of a variety of pedigreed monitoring strategies that could improve and prolong the use of scarce plant resistance alleles in rice and other plants.