Classification of Poultry Meat Cuts Based on Approach of Untargeted Lipidomic Analysis and Advanced Chemometrics

Significance of the topic

Poultry meat is a primary source of dietary lipids, and consumer demand for lean, health‐oriented products places emphasis on understanding fat composition at the cut level. Determining the lipid profiles of chicken breast versus thigh supports nutritional guidance, product differentiation, and authenticity verification in the food industry.

Study Objectives and Overview

This investigation aimed to apply an untargeted lipidomic workflow coupled with advanced chemometric approaches to compare and classify the lipid composition of chicken breast and thigh cuts. Two main goals were pursued: (1) to identify compositional differences in saturated and unsaturated lipid species, and (2) to construct statistical models for reliable discrimination of the two meat cuts.

Methodology and Instrumentation

Sample Preparation and Extraction:

• Approximately 100 mg of lyophilized poultry muscle (breast or thigh) were ground and extracted using a biphasic MTBE:MeOH:H₂O protocol.
• The MTBE layer was collected, evaporated under nitrogen, and reconstituted in IPA:MeOH (2:1, v/v).

Analytical Platform:

• Reversed‐phase UPLC coupled to trapped ion mobility spectrometry quadrupole time‐of‐flight mass spectrometry (RP-UPLC-TIMS-TOF-MS) in positive ionization mode with PASEF technology.
• Column: Thermo Acclaim RSLC C18 (2.1 × 100 mm, 2.2 μm); mobile phases of ACN:H₂O (65:35) and ACN:IPA (15:85) with 10 mM ammonium formate and 0.1% formic acid.
• Scan range m/z 150–1350, source temperature 200 °C, flow rate 0.25 mL/min.

Data Treatment:

• MetaboScape 2023 internal lipid annotation tool and LipidBlast spectral library were used to annotate mass features.
• Chemometric analysis included hierarchical clustering (HCA) and orthogonal partial least squares discriminant analysis (OPLS-DA), with VIP extraction and permutation tests to validate model robustness.

Main Results and Discussion

Over 1 700 mass features were detected and more than 200 lipid molecules annotated; 170 lipids were confidently characterized. Both HCA and OPLS-DA models achieved clear separation of breast and thigh samples, with zero misclassification in validation. Seventy‐four compounds exhibited VIP scores above 1.0, including key phosphatidylcholines, triglycerides, diglycerides, and sphingomyelins. Nutritionally, breast meat contained approximately 25% less saturated fat and displayed higher PUFA/SFA and MUFA/SFA ratios compared to thigh.

Benefits and Practical Applications

The described workflow provides a robust, high-throughput approach for meat cut authentication and nutritional profiling. It can support quality control, label compliance, and product differentiation in poultry processing and regulatory settings.

Future Trends and Potential Applications

Emerging directions include integration of targeted quantitation methods for absolute lipid concentrations, expansion of spectral libraries to cover minor lipid classes, application of machine learning algorithms for enhanced classification, and on‐line monitoring systems in industrial processing lines for real‐time quality assurance.

Conclusion

The combination of untargeted lipidomics via RP-UPLC-TIMS-TOF-MS and advanced chemometrics reliably discriminates chicken breast from thigh based on lipid composition, highlights nutritional distinctions, and offers a powerful tool for meat authenticity and quality assessment.