Effects of data structure and selection on estimated inbreeding depression in experimental Tribolium castaneum lines

Inbreeding depression was estimated for four experimental Tribolium castaneum lines. Each line, containing approximately 7000 animals, was selected for 16 generations either randomly (control), on pupae weight (PWT), on family size (FST) or on an index containing both PWT and FST. The inbreeding trend was 0.9, 0.5, 0.5 and 0.4% inbreeding per generation in PWT-selected, FST-selected, index-selected, and control line, respectively. The model used to estimate the inbreeding depression included a linear regression on individual inbreeding coefficients, and random additive genetic effects. Using all the performance and pedigree data, estimated inbreeding depressions in the control line were -0.13 (SE = 0.16; #) and -8.50 (SE = 2.66; mu g) per 1% inbreeding for FST and PWT, respectively. Using only performance data of the latest generation in the control line, the estimated inbreeding depressions changed considerably: -0.17 (SE = 0.82) and -37.4 (SE = 11.9) for FST and PWT, respectively. Estimated inbreeding depression for FST in the FST-selected line was -0.40 (SE = 0.31). Inbreeding depression for PWT in the PWT-selected line was 21.6 (SE = 25.8). This study indicates that estimating inbreeding depression might best be based on the performance data of animals with an equal and sufficiently-large number of ancestral generations known.