Chemotaxonomic Classification of Annonaceae Species From Herbarium Samples Using LC-HRMS, Supported by Multivariate Curve Resolution Applied to Regions of Interest (ROIMCR)

Chemotaxonomic approaches provide a valuable complement to morphological analysis and DNA barcoding techniques for classifying herbarium specimens, particularly in plant families where closely related species exhibit overlapping diagnostic traits. In this study, liquid chromatography–electrospray high-resolution mass spectrometry (LC–ESI–HRMS) data from 53 Annonaceae herbarium leaf samples were processed using the ROIMCR workflow, which integrates region-of-interest (ROI) filtering with multivariate curve resolution-alternating least squares (MCR-ALS). The resulting component areas were used to construct a hierarchical clustering that largely recovered species-level groupings, despite collection dates spanning more than two decades. This chemotaxonomic analysis corroborated most herbarium identifications, highlighted probable misassignments, and enabled the classification of previously unidentified specimens. Discriminant component analysis revealed that aporphine alkaloids and acetogenins constituted the principal chemical markers driving the separation of Annonaceae genera and species. The extension of the approach to bark and flower extracts showed that leaves and bark share many metabolites but differ markedly in relative abundances, which complicates building a unified, organ-independent classification. Overall, this work demonstrates that ROIMCR-based chemotaxonomy provides a robust LC–MS-compatible framework to exploit archived herbarium material and to strengthen species assignment in metabolite-rich plant families such as Annonaceae.