Methods of analysis and the estimation of the genetic parameters from a diallel set of crosses

• Published in: Heredity 1984-01, Vol.52 (2), p.243-253
• Main Authors: Pooni, H S, Jinks, J L, Singh, R K
• Format: Article
• Language: English
• Subjects: Biological and medical sciences; Biomedical and Life Sciences; Biomedicine; Classical genetics, quantitative genetics, hybrids; Cytogenetics; Ecology; Evolutionary Biology; Fundamental and applied biological sciences. Psychology; Genetics of eukaryotes. Biological and molecular evolution; Human Genetics; Methods, theories and miscellaneous; original-article; Plant Genetics and Genomics
• ISSN: 0018-067X; 1365-2540
• DOI: 10.1038/hdy.1984.26

Summary The relationships between the estimates of general (σ 2 g ) and specific (σ 2 s ) combining abilities obtained from the four methods of diallel analysis described by Griffing (1956) and estimates of D R and H R , the additive and dominance components of variability, are compared both theoretically and experimentally for model II situations. Theoretically they are compared by deriving the expectations of the variance components for the general and specific combining abilities for each method in terms of the genetical parameters and Experimentally they are compared by extracting data from a diallel set of crosses between a random sample of 29F 9 families derived from the F 2 of the cross of varieties 1 and 5 of Nicottana rustica by pedigree inbreeding. The theoretical results show and the experimental results confirm that the genetical expectations of σ 2 g and σ 2 s for method 1 are identical with the general definitions of ¼ D R and ¼ H R . This method, therefore, gives consistent estimates of D R and H R in all situations. In contrast, methods 2, 3 arid 4 give close approximations to ¼ D R and ¼ H R only when estimates from diallels involve large numbers of parents. The limitations of diallel analysis as a source of estimates of genetically defined as opposed to statistically defined parameters are discussed.