Análise de componentes principais no estudo de medidas de ultrassom in vivo em animais da raça nelore

Abstract

Carcass ultrasound is a valuable tool for evaluating traits associated with meat quality, offering accurate, low-cost, and rapid in vivo assessments. This technique enables the measurement of key carcass characteristics, including ribeye area (REA), subcutaneous fat thickness (SFT), and marbling score—typically assessed between the 12th and 13th ribs using real-time ultrasound. These traits, may be influenced by environmental factors such as birth year and month, sex, dam age at calving, and the animal's age. This study aimed to investigate the influence of environmental factors on ultrasound carcass traits using mixed factor analysis. The dataset included approximately 3,250 ultrasound records collected at weaning from male and female Nellore cattle raised on a farm in Bela Vista, MS, Brazil, between 2018 and 2022. Multivariate analysis was performed using the mixed factor analysis (MFA) method in the R statistical software, allowing for the exploration of interrelationships among multiple variables and graphical interpretation for decision-making. The average body weight was 194.7 kg; REA averaged 41.73 cm²; SFT was 2.63 mm; and the marbling score averaged 2.98. Principal component analysis (PCA) reduced 13 variables to five principal components, which explained 71.87% of the total variance. Component 1 was primarily associated with weight, age, REA, and SFT; component 2 was related to sex; and component 3 to birth year. Marbling and birth month were strongly associated with component 5, indicating potential seasonal effects. Graphical outputs revealed groupings among individuals based on multiple factors. While no clear separation was observed by sex or dam age, birth year, birth month, and calf age significantly influenced carcass traits. These findings highlight the importance of strategic management during the calving season and suggest promising potential for genetic selection.