Plant & Animal Genome VI Conference Plant & Animal Genome VI Conference
W17
Two simulations were conducted to evaluate the probability
of false error rate and detection (both location and size)
of QTL for a random mating population and a designed QTL
population where sires were heterozygous for QTL. Two
hundred families were generated for each heritability (.10,
.25 and .40) and half-sib family size (50, 150 and 250).
Individual genotypes (10 chromosomes, 1180 cM, 38 markers,
5 QTL of .5 to 3.0 additive genetic SD units) and phenotypes
(based on QTL genotype, additional polygenic variation not
accounted for by QTL and environmental effects) were
simulated. Data were analyzed with ANIMAP, an interval
mapping algorithm. Both populations resulted in similar
false error rates and power to detect QTL. False error rate
per 118 cM length was dependent upon family size and
significance level (LOD score). Families of 250 progeny had
a false error rate of .63 per 100 simulations with a minimum
LOD score of 2.0, families of 50 progeny had a false error
rate of .43 per 100 simulations with a minimum LOD score of
2.5. The power to detect QTL in the correct marker interval
ranged from .005 (QTL=.5 additive genetic SD units) to .61
(QTL=3.0 additive genetic SD units) in the large family.
There are several technologies available or being
developed to screen potential sires for heterozygosity at a
major gene or QTL based on pedigree and phenotypic data.
FINDGENE, a technology developed in Australia, has been used
to evaluate both Angus and Limousin databases, and sires have
been identified that have a high probability of segregating
major genes for carcass traits. In addition, another method
of evaluating parental heterozygosity using within-family
additive genetic variance will be presented.