power and parameter estimation of complex segregation analysis under a finite locus model权力和复杂的分离分析的参数估计在一个有限的轨迹模型.pdfVIP

Original article Power and parameter estimation of complex segregation analysis under a finite locus model P Uimari, BW Kennedy JCM Dekkers Department of Animal and Poultry Science, Centre for Genetic Improvement of Livestock, University of Guelph, Guelph, ON N1G 2W8, Canada (Received 20 October 1995; accepted 7 May 1996) Summary - Power and parameter estimation of segregation analysis was investigated for independent nucleus family data on a quantitative trait generated under a finite locus model and under a mixed model. For the finite locus model, gene effects at ten loci were generated from a geometric series. Additionally, linkage between a major locus and other loci was considered. Two different methods of segregation analysis were compared: a mixed model and a finite polygenic mixed model. Both statistical methods gave similar power to detect a major gene and estimates of parameters. An exception was a situation where two major loci had an equal effect on phenotype: the mixed model had a higher power than the finite polygenic mixed model, but estimates of the parameters from the mixed model were more biased than estimates from the finite polygenic mixed model. Segregation analysis was more powerful in detecting a major gene when data were generated under the finite locus model than under the mixed model. When a major gene was linked to another gene, a major gene was more difficult to detect than without such linkage. Segregation of two major genes created biased estimates. Bias increased with linkage when parents were not a random sample from a population in linkage equilibrium. parameter estimation / power / major gene / segregation analysis Résumé - Puissance et estimation des paramètres dans l’analyse de ségrégation com- plexe avec un modèle à nombre fini de l