Near infrared spectroscopy (NIRS): a new technology to predict intramuscular fat in lamb meat

• Grant number: 19/12125-9
• Support Opportunities: Scholarships abroad - Research Internship - Doctorate
• Start date: August 30, 2019
• End date: February 28, 2020
• Field of knowledge: Agronomical Sciences - Animal Husbandry
• Principal Investigator: Eduardo Francisquine Delgado
• Grantee: Giuliana Micai de Oliveira
• Supervisor: David Laurence Hopkins
• Host Institution: Escola Superior de Agricultura Luiz de Queiroz (ESALQ). Universidade de São Paulo (USP). Piracicaba , SP, Brazil
• Institution abroad: Cowra Agricultural Research and Advisory Station, Australia
• Associated to the scholarship: 18/01780-3 - Implications of pregnant sheep nutrition on myofibrils and progeny blood parameters, BP.DR

Abstract

To improve performance of animals after birth, consideration should be given to maternal nutrition throughout pregnancy, and this strategy of improvement involves aspects related to what is called "fetal programming". A bad maternal nutrition during the gestational period can lead to problems in myogenesis and adipogenesis. The amount and distribution of intramuscular fat (IMF) plays an important role in the quality of meat foods, because they can affect characteristics such as taste, juiciness, texture and appearance. Thinking about that, the use of near infrared visible spectroscopy (NIRS) becomes interesting for monitoring, quality control and general analysis, which can provide an objective, repetitive, fast, accurate and non-destructive evaluation method of lamb meat. With this, the goal of the internship will be to collaborate with the validation of the NIR spectroscopy analysis to predict the IMF. Measurements will be conducted at approximately 20 minutes and 24 hours post mortem, and these will be compared. Data will be collected on both the loin and the topside, and two NIRS devices will be used for this. After that, a sub-section of 25-30 g will be removed for analysis of the IMF by the Soxhlet method. The spectra will be converted to absorbance, and then the outlying spectra will be removed. Partial least squares analysis will be done and will be completed using the number of latent variables based on the optimal reduction of the mean squared error of cross validation. To test the robustness of the model the data set will be divided into calibration and validation subsets. The calibration model will be used to predict the IMF values of the muscles. (AU)