Premium
Expected Genetic Advances from Three Simulated Selection Schemes 1
Author(s) -
Frey K. J.
Publication year - 1968
Publication title -
crop science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.76
H-Index - 147
eISSN - 1435-0653
pISSN - 0011-183X
DOI - 10.2135/cropsci1968.0011183x000800020030x
Subject(s) - heritability , replicate , selection (genetic algorithm) , biology , trait , grain yield , genetic gain , statistics , genetic correlation , mathematics , agronomy , genetic variation , genetics , computer science , gene , machine learning , programming language
In an eight‐replicate oat experiment with 78 F 2 ‐derived and 234 F 3 ‐derived lines (3 from each F 2 ), I measured grain yield, heading date and plant height, and in a four‐replicate barley experiment with 90 F 2 ‐derived and 270 F 3 ‐derived lines (3 from each F 2 ), I measured grain yield, plant height and weight per volume. I conducted three simulated selection experiments using portions of these data. The resource input for each selection experiment was sufficient to evaluate 78 oat lines in eight replicates and 90 barley lines in four replicates. In one case (random F 3 system), random F 3 ‐derived lines were chosen without regard to the F 2 's from which they were progenies. In a second case (stratified F, system), stratified F 3 ‐derived lines (one F 3 ‐derived line from each F 2 ) were chosen, and in the third case (tandem system), group of derived lines were evaluated and the best 2056 for the trait under selection, were saved. Families of F 3 ‐derived lines (three per family) were tested for each selected F 2 . Expected genetic gatns were calculated from the three selection systems for the six trait‐species combinations. In five of six trait‐species combinations, the tandem system produced greater expected genetic gains than either the random or stratified F 3 systems. The latter two produced about equal selection gains. The degree of superiority of the tandem system was related to the heritability percentage of the trait in question. With low heritability percentage, the expected genetic gain from the tandem method was more superior than when the heritability percentage was high. With barley, 40 to 50% of the expected genetic gain from the tandem system resulted from the F 3 ‐derlved line test, but most of this gain was due to exploiting genotypic variance among selected F 2 families. With oats much of the expected genetic gain from the second phase of the tandem system was due to exploiting genotypic variances among F 3 ‐derived lines within F 2 families.