Evaluation of biometrical methods for estimating the number of genes : 1. Effect of sample size.

The effect of sample size on estimating the number of genes by the inbred-backcross and genotype assay procedures was investigated. Modifications were proposed for each procedure. Ninety-five percent confidence intervals for estimated numbers of genes and the minimum sample size required to discriminate between various genetic hypotheses were calculated for both procedures. Sample size had a greater impact on the estimation of gene number by the genotype assay procedure than by the inbred-backcross procedure, especially for small sample sizes. For the inbred-backcross procedure, the optimal number of backcrosses varied with the number of genes. Estimates of the number of genes are theoretically less reliable when estimated by the genotype assay procedure than by the inbred-backcross procedure, and are sensitive to the choice of assay generation. Generally, the inbredbackcross procedure is preferred. Even with the fulfillment of all genetic assumptions for each method and absence of error in measuring genotypic values, substantial upward or downward biases in the estimates of the number of genes are expected from both the inbred-backcross and the genotype assay procedures.