Search

The objective of the present study is to evaluate the potential use of FT-NIRS for predicting intramuscular fat (IMF) and fatty acid groups (MUFA; PUFA; PUFA n-3, PUFA n-6; SFA) on pig grounded muscles. The research considered 160 fresh samples of Longissimus collected from 12 European local pig breeds (TREASURE* project). For every sample, lipids were extracted from IMF and fatty acid profile was determinate by gas chromatography. Two aliquots of each sample were scanned using FT-NIRS Antaris II model. Mathematical pre-treatments (multiplicative scatter correction, 1st and 2nd derivate) were applied and outliers’ spectra were identified and removed when necessary. Partial least square regression was used on the average spectrum and the models validated using an external data set. Results are evaluated in terms of coefficient of regression and root mean square errors in calibration (R2- RMSE) and validation (Rp2-RMSEP). As expected, the best results were obtained for IMF with R2 higher than 0.99 and RMSE lower than 0.2. Unsaturated fatty acids, probably due to the absorption of the cis double bond in a specific region of near infrared spectra, obtain acceptable R2 (0.89 for MUFA and 0.75 for PUFA n-3 and PUFA n-6). SFA achieved a R2 of 0.81 that is lower than values reported in other studies probably because of the large variability of genotypes used. The validation models achieved both lower coefficients of determination and higher RMSEP than the calibration models; however, R2 differences between calibration and validation were smaller than 5%, except for SFA. Hence, the FT-NIRS seems promising to estimate the principal parameters of fatty acid groups on muscle samples from different European autochthonous pig breeds. Inclusion of other samples can improve the accuracy and the robustness of the models, especially considering the high variability of the samples. *Funded by European Union’s H2020 program (grant agreement No 634476).