Multilocus epistasis, linkage, and genetic variance in breeding populations with few parents

The preferential use of only two inbreds in farming breeding populations restricts the amount of variability expected in a cross and may ultimately lead to loss of genetic variance. The objectives of this study were: 1) to test if genetic variance in maize (Zea mays L.) decreases as the number of parents (N) decreases as expected from an additive genetic model, and 2) to determine how multilocus epistasis and linkage affect the loss of genetic variation in populations established from small N. Eight F3 families derived from the same single cross, were intermated in a hierarchical manner to form populations with N=1, 2, 4, or 8 parents. Test cross genetic variance (V sub TC) generally decreased with N, although differences observed at N=2, 4, and 8 were mostly insignificant. In particular, grain yield V sub TC at N=4 was not significantly different from the V sub TC in the base populations, indicating a genetic variance in excess of what is predicted by an additive model. In metabolic control simulation experiments, the average ratio between genetic variance in the progeny population (V sub N) and the base population (V sub B) was lower under metabolic flux epistasis than under additivity, indicating that epistasis hastens the decline in genetic variance due to small N. In contrast, the 95th (Delta 95 percent) percentile difference between the estimated V sub N/V sub B and the expected V sub N/V sub B was higher with epistasis than with pure additivity across the different levels of N and number of loci. Linkage had little effect on V sub N/V sub B, whereas it increased Delta 95 percent under both additivity and epistasis. The results implied that non-additive gene effects help maintain genetic variance in small populations in elite maize inbreds, and that metabolic flux epistasis leads to greater variability resulting in some populations having V sub N/V sub B that was larger than expected.