Identification of potential quality markers in Indonesia’s Arabica specialty coffee using GC/MS-based metabolomics approach

Introduction The cupping test is a widely used method for quality assessment of Arabica coffee. However, the cupping test is limited by the low number of certified panelists and the low throughput. Therefore, an analytical-based quality assessment may be a promising tool to complement the cupping test. A present, there is no report investigating quality marker candidates, focusing only on “specialty” grade Arabica coffee from Indonesia. Objective This study identified the potential quality marker(s) in Arabica Specialty coffee at different stages (green beans, roasted beans, and brewed coffee. Methods The metabolite profiles of ten different Arabica specialty-grade coffees were analyzed with different cup scores using gas chromatography–mass spectrometry (GC/MS). From the ten samples, green coffee beans, roasted coffee beans, and brewed coffee were selected. In addition, an orthogonal projection to latent structure (OPLS) regression analysis was conducted to obtain a potential quality marker based on the variable importance in projection (VIP). The potential quality marker(s) were validated by GC/MS metabolome profiling and OPLS analysis of different sets of samples consisting of 35 Arabica specialty-grade coffee samples. Results In Arabica coffee samples, the OPLS model of the three stages showed galactinol to have a high VIP score. Galactinol showed a consistent positive correlation with cup scores at all stages of coffee production (green beans, roasted beans, and brewed coffee). The correlation suggests galactinol is a potential quality marker after further validation using different samples. Conclusion GC/MS combined with OPLS regression analysis suggested galactinol as a quality marker and provide an early screening method for Arabica coffee quality that complements the cupping test performed by certified panelists.