The use of real-time-ultrasound to predict genetic attributes of body composition traits in live beef cattle

Thumbnail Image
Date
1996
Authors
Izquierdo Cebrián, Mercedes
Major Professor
Advisor
Doyle Edward Wilson
Committee Member
Journal Title
Journal ISSN
Volume Title
Publisher
Altmetrics
Authors
Research Projects
Organizational Units
Organizational Unit
Animal Science

The Department of Animal Science originally concerned itself with teaching the selection, breeding, feeding and care of livestock. Today it continues this study of the symbiotic relationship between animals and humans, with practical focuses on agribusiness, science, and animal management.

History
The Department of Animal Husbandry was established in 1898. The name of the department was changed to the Department of Animal Science in 1962. The Department of Poultry Science was merged into the department in 1971.

Historical Names

Journal Issue
Is Version Of
Versions
Series
Department
Abstract

Body composition traits (BCT) records from 1003 beef cattle were collected from 1991-1995. Data collected in the longissimus dorsi (ld) muscle included, 12th-13th rib carcass fat thickness (FAT), USDA marbling score (MS) and chemical percentage of intramuscular fat (PIFAT). Before slaughter, a cross-sectional ultrasound ld image between the 12th-13th ribs was collected to calculate ultrasound fat thickness (UFAT). A longitudinal ld image across 11th-12th-13th ribs was used to calculate image analysis parameter including: histogram, Fourier and texture. Multiple regression and cluster analysis were used to develop prediction models for ultrasound percentage intramuscular fat (UPIFAT) from the image parameters. An independent data set was used to validate the prediction models. BCT genetic variance and covariance parameters were computed at age- and weight-constant end points for bulls, steers, and both sexes combined using computer algorithms of MTDFREML for variance component estimation. Sire breeding values (BV) were ranked for ultrasound traits and corresponding carcass traits. Prediction accuracy for PIFAT values ranging from.5% to 13% resulted in a robust and unbiased model with a root mean square error (RMSE) of 1.43% and a coefficient of determination (R-square) of.59. This model included exclusively image analysis parameters. For actual PIFAT values between.5% and 6% PIFAT can be predicted with an average error of ±.9%. This PIFAT interval includes the majority of the scanned animals. The use of cluster analysis slightly reduced RMSE in the lower PIFAT classes to 1.13%. Genetic parameters were significantly different for bulls and steers. Genetic parameters adjusted to a weight-constant end point were smaller than those adjusted to an age-constant end point. The genetic correlations between FAT and UFAT and between PIFAT and UPIFAT indicate that these paired traits are controlled by the same genes. The correlation between PIFAT sire BV and UPIFAT sire BV increases when the prediction error variance is reduced by increasing the number of progeny per sire. It is concluded that UPIFAT can be used to accurately rank sires BV for PIFAT by using progeny testing. When there are more than 8 progeny per sire, RTU determined BV are correlated with carcass determined BV at a level of.80.

Comments
Description
Keywords
Citation
Source
Copyright
Mon Jan 01 00:00:00 UTC 1996