Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.
Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.
Generation means estimation of unbiased genetic effects from five cycles of replicated S1 and reciprocal full-sib recurrent selection
Abstract
Evaluations of recurrent selection procedures are needed to determine the best procedures for the improvement of plant populations. When small effective population sizes are used in selection programs, random genetic drift influences the performance of populations and the effects of drift and selection can be confounded in progress evaluations. Generation means models are designed to estimate drift effects so confounding of selection and drift effects can be eliminated. The objectives of this study were to modify present generation means theory for replicated selection populations to obtain drift free estimates of selection effects, and combine large, unbalanced data sets for an overall evaluation of the S1 and RFS selection programs. Model equations indicate that the previous term used to estimate drift effects is actually composed of quadratic gene frequency changes due to both drift and selection. The sign of the quadratic gene frequency changes due to selection may be either positive or negative and thus previous estimates of drift may be biased either upward or downward. Linear combinations of genetic effects relating to the changes per cycle due to selection for populations per se, selfed, and crossed were calculated. S1 selected materials were superior to RFS materials when evaluated as selfed populations with the S1 method increasing at a rate of 4.633 q/ha/cycle and the RFS method increasing at a rate of 2.458 q/ha/cycle + 0.306 q/ha/cycle$\sp2$ for the selection index. As populations per se, the S1 method improved the selection index by 5.221 q/ha/cycle and the RFS method improved by 2.653 q/ha/cycle + 0.153 q/ha/cycle$\sp2$. Population cross performance was 4.663 q/ha/cycle + 0.268 q/ha/cycle$\sp2$ for S1 material and 3.383 q/ha/cycle + 0.612 q/ha/cycle$\sp2$ for RFS material. Gains in performance of S1 populations per se, selfed, and crossed were due primarily to cumulative changes in the frequency of alleles with additive effects. Changes in RFS populations were most evident in population crosses and were a result of cumulative changes in the frequency of alleles with additive effects and increases in heterosis.
Subject Area
Agronomy|Genetics
Recommended Citation
Tragesser, Scott Louis, "Generation means estimation of unbiased genetic effects from five cycles of replicated S1 and reciprocal full-sib recurrent selection" (1991). ETD collection for University of Nebraska-Lincoln. AAI9208117.
https://digitalcommons.unl.edu/dissertations/AAI9208117