Posterior probabilities for Beta-mannosidosis in Salers cattle

Abstract

A total of 5,743 records of plasma P-mannosidase activity from Salers cattle including concomitant pedigree information and main effects of gender, age, month of testing, lactation, pregnancy, body condition score, and breed composition were incorporated into models fit by maximum likelihood to determine genotype and posterior probabilities for P- mannosidosis under a diallelic model of inheritance. The objective of this study was to develop a screening test for the detection of the lethal P-mannosidosis allele in Salers cattle through the development of models with expanded capabilities for genotype discrimination. Reduced models including restrictions on either main effects or pedigree information were compared to the basic model including the most complete information on main effects and pedigree. Estimation of polygenic effects was attempted by the use of the MTDFREML program. The basic model was considered to be the most appropriate for parameter estimation and genotype determination for P-mannosidosis. Correction for sample ascertainment was achieved through the incorporation of three generations of pedigree information. Gene and carrier genotype frequencies were 0.0713 and 0.1426, respectively. Estimates of the genotype major phenotypic variances were 598.21 and 750.02 (units/MI)2 for Aa and AA, respectively. The difference in fitness between genotypes was not biologically significant. In general, main effects were more significant for homozygotes. Age and gender were the effects with the greatest influence on P-mannosidase activity. Enzyme activity tended to be higher during the winter months. Pregnancy had a negative effect on enzyme activity for females, whereas lactation had a small negative effect on carriers but a large positive effect on homozygotes. Linear body condition score had a negative effect on all animals except homozygous males, and the quadratic term for this effect was not statistically significant. Breed composition was most significant for 1/2 Salers. The polygenic effects model did not appropriately partition the major gene and polygene effects on P-mannosidase activity. This study produced a method for the computation of genotype posterior probabilities that represents an efficient and cost effective means for the control and elimination of the lethal P- mannosidosis allele from the U.S. Salers cattle population.