Submitted to: North American Alfalfa Improvement Conference
Publication Type: Abstract Only
Publication Acceptance Date: 5/6/2014
Publication Date: 5/15/2014
Citation: Riday, H., Smith, M.A., Peel, M. 2014. Modeling forage legume polycross pollen-parent gamete contributions to progeny based on empirical measurements with implications for effective polycross size [abstract]. North American Alfalfa Improvement Conference. Paper No. 15.
Technical Abstract: Random mating is a fundamental assumption in quantitative genetic theory and assumes pollen-parents contribute uniform gamete numbers to the progeny generation; polycross-size-expected-inbreeding rests on this assumption. This study empirically determined polycross pollen-parent gamete frequency contribution to the progeny generation and developed a model for these frequencies that depends on polycross pollen-parent number. Ten isolated forage legume polycrosses were evaluated ranging from 9 to 94 polycross pollen-parents, in three species [alfalfa (only outcross progeny considered), kura clover and red clover], pollinated with bumble or leafcutter bees. The Pioneer, Alforex, and KU09B polycrosses included clonally replicated parents. Parents and 321 to 1704 progeny per polycross were genotyped and paternity tested (seed-parent was known). Individual pollen-parent frequency contributions to the progeny generation were calculated for each polycross. Pollen-parent frequencies and polycross size were standardized so comparisons between different-sized polycrosses could be made. In no polycross was a uniform pollen-parent contribution to progeny observed, indicating that a standard assumption of random mating is never observed. Weibull distributions fit best with observed pollen-parent frequencies; specifically an inverse Weibull cumulative distribution function fit the standardized combined across polycross pollen-parent frequencies with an R-sq. = 0.953. A model was developed to predict pollen-parent frequency distribution of gamete contribution to the progeny generation given the number of pollen-parents in a polycross. This model was then applied to the simple situation of a polycross with uniform seed-parent gamete contribution to progeny (i.e., equally bulked seed-parent halfsib seed) and the modeled expected pollen-parent gamete contribution to progeny. Effective polycross size was reduced in a linear fashion (N-Effective = 0.9N-Actual + 0.485, r-sq. = 1), resulting in a 7-10% effective polycross size reduction for polycross sizes ranging from 15 to 100. Further implications of an inverse Weibull cumulative distribution function pollen-parent gamete contribution frequency distribution to the progeny generation should be examined in the context of other known perturbations to classic random-mating such as distance dependency and fecundity.