The authentication of Yanchi tan lamb based on lipidomic combined with particle swarm optimization-back propagation neural network
This study successfully combined widely targeted lipidomic with a back propagation (BP) neural network optimized based on a particle swarm algorithm to identify the authenticity of Yanchi Tan lamb. An electronic nose and gas chromatography–olfactometry-mass spectrometry (GC-O-MS) were used to explor...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-12-01
|
| Series: | Food Chemistry: X |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2590157524009192 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | This study successfully combined widely targeted lipidomic with a back propagation (BP) neural network optimized based on a particle swarm algorithm to identify the authenticity of Yanchi Tan lamb. An electronic nose and gas chromatography–olfactometry-mass spectrometry (GC-O-MS) were used to explore the flavor differences in Tan lamb from various regions. Among the 17 identified volatile compounds, 16 showed significant regional differences (p < 0.05). Lipidomic identified 1080 molecules across 41 lipid classes, with 11 lipids, including Carnitine 15:0, Carnitine 17:1, and Carnitine C8:1-OH, serving as potential markers for Yanchi Tan lamb. In addition, a stepwise linear discriminant model and three types of BP neural networks were used to identify the origin of Tan lamb. The results showed that particle swarm optimization-back propagation (PSO-BP) neural network had the best prediction effect, with 100 % prediction accuracy in both the training and test sets. The established PSO-BP model was able to achieve effective discrimination between Yanchi and non-Yanchi Tan lamb. These results provide a comprehensive perspective on the discrimination of Yanchi Tan lambs and improve the understanding of Tan lamb flavor and lipid composition in relation to origin. |
|---|---|
| ISSN: | 2590-1575 |