Amino acid profiles and quantitative structure–property relationship models as markers for Merlot and Torrontés wines

Quantitative structure–property relationships (QSPRs) were applied to the aminograms obtained by HPLC in our laboratories for Torrontés and Merlot wines. Dragon theoretical descriptors were derived for a set of optimized amino acid structures with the purpose of establishing QSPR models. The statistical Replacement Method was used for designing the best multi-parametric linear regression models, which included structural features selected from a pool containing 1497 constitutional, topological, geometrical or electronic molecular descriptors. Predicted QSPR results were in good agreement with experimental amino acid profiles. The developed QSPR approach showed to be of practical value for distinguishing each wine varietal, and for calculating experimentally non-available amino acid concentrations of Torrontés and Merlot wines. It was also useful for assessing wine authenticity; the models were especially suitable for Merlot and Torrontés wines.

► Aminograms account for fingerprints of free amino acids and ammonia in wines. ► Linear QSPR models for predicting aminograms in Merlot and Torrontés wines. ► Molecular structure represented through Dragon molecular descriptors. ► Validation of the QSPR models with an external test set and Cross-Validation. ► These are among the first QSPR for predicting aminograms of wines.