Differentiation between house and cave edible bird's nests by chemometric analysis of amino acid composition data

- Amino acid profile of house and cave edible bird's nests were analyzed by GC-MS.
- Distinct clusters based on edible bird's nest sample types were observed by PCA.
- Predictive model with high predictive ability has been constructed using OPLS-DA.
- The model built was a promising tool for identifying the type of new unknown samples.
- Discriminatory markers suggested by the model were TYR and GLU.

A total of 60 pieces of house and cave edible bird's nest (EBN) samples were randomly collected from different locations in Malaysia and Indonesia. Amino acid compositions of the EBN samples were determined by gas chromatography-mass spectrometry (GC-MS). Data obtained was analyzed by Pearson Correlation analysis, Principal Component analysis (PCA) and Orthogonal Partial Least Square-Discriminant analysis (OPLS-DA). There were highly significant different correlations seen among amino acids in house- and cave-EBN samples. The model constructed by OPLS-DA was found to be a promising tool with high predictive power of 76.1%. Robustness of the model was validated and blind test samples were correctly assigned to their respective cluster. Tyrosine (TYR) and glutamic acid (GLU) were proposed as the promising markers for differentiating between house and cave EBN.