Multivariate analyses of meat quality data as a tool to validate the macroscopic classification of Pectoralis major muscles affected by Wooden Breast myopathy

The present study is part of the PACHOL4 project seeking to deepen the knowledge about the molecular causes of the Wooden Breast (WB) defect, which affects the Pectoralis major muscles (PMs) of fast-growing chickens and severely compromises the main quality traits of affected meat. The primary goal of this project was to accurately classify the PMs as affected and not affected cases, to ultimately identify the samples that better represent the group they belong to, and, thus, select those to be considered for further analyses foreseen by the project. To this purpose, a total of 180 PMs belonging to Ross 308 chickens (males 45-d old, 3.2 kg of live weight; DWG>85g) were collected at a commercial abattoir 3 h post-mortem and classified according to their macroscopic features as normal (NORM; n=90/group) and abnormal (WB; n=90/group). Quality traits (pHu, color, drip and cooking losses, and shear force) were evaluated. The WB effect on the PMs quality traits was tested by Student t-test in R (v.4.3.2). Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) were performed in R using meat quality data. Then, based on the PCA and HCA results, a subset of samples (n=70; 35/group) were selected and used to test the WB effect on meat quality traits (Student t-test; R v.4.3.2) of samples chosen for further investigation. Considering the overall dataset (n=180), meat quality evaluation evidenced significant differences (P<.05) between NORM and WB concerning the pHu, redness (a*), cooking loss, and shear force. Based on PCA, the parameters of pHu, lightness (L*), cooking loss, and shear force were identified as potentially useful for sample classification. HCA showed sample clustering partially resembling the macroscopic classification implemented at the processing plant and allowed us to select an arbitrary subset of samples (n=70; 35/group). Unlike the results obtained considering the whole dataset, the effect of WB on meat quality traits of selected samples evidenced differences (P<.05) between NORM and WB concerning L* and yellowness (b*) but not a*. Notably, these results seemed to better overlap the evidence already reported in the literature, which mainly observed increased L* and b* values in WB, thus corroborating a better representation of the group they were classified into and supporting their consideration for subsequent analyses. Research supported by NextGenerationEU, National Grant PRIN2022 (Prot.n.2022EPWEPW).