Abstract
Metabolomics serve as a powerful tool for understanding the biochemical diversity of plants and its relationship with ecological characteristics [1]. Plant metabolic signatures reflect their adaptive strategies, which influence their functional groups, growth forms, and phenological patterns [2]. Recent advancements in machine learning have enhanced the classification and prediction of biological traits derived from complex biochemical datasets [3]. However, the use of metabolomic data for predicting ecological and evolutionary trends in plants remains mostly underexplored. We analyzed the metabolomic profiles of 74 plant species from 38 plant families grown under natural conditions. Different machine learning models were used and enabled us to identify metabolic signatures associated with plant functional groups, growth forms, and phenology with high accuracy, revealing significant correlations between metabolites and botanical traits. Our results demonstrate that metabolomic profiling, along with machine learning, can effectively predict ecological and evolutionary trends in plants. This approach provides a new framework for investigating plant adaptation and ecological strategies.References1. Fiehn O. Metabolomics--the link between genotypes and phenotypes. Plant Mol Biol. 2002 Jan;48(1–2):155–71.2. Petrén H, Anaia RA, Aragam KS, Bräutigam A, Eckert S, Heinen R, et al. Understanding the chemodiversity of plants: Quantification, variation and ecological function. Ecological Monographs. 2024 Nov 1;94(4):e1635.3. Dussarrat T, Prigent S, Latorre C, Bernillon S, Flandin A, Díaz FP, et al. Predictive metabolomics of multiple Atacama plant species unveils a core set of generic metabolites for extreme climate resilience. New Phytologist. 2022;234(5):1614–28.