A machine learning approach to study plant functional trait divergence.

ABSTRACT:

Premise

Plant functional traits are often used to describe the spectra of ecological strategies used by different species. Here, we demonstrate a machine learning approach for identifying the traits that contribute most to interspecific phenotypic divergence in a multivariate trait space.

Methods

Descriptive and predictive machine learning approaches were applied to trait data for the genus Helianthus, including random forest and gradient boosting machine classifiers and recursive feature elimination. These approaches were applied at the genus level as well as within each of the three major clades within the genus to examine the variability in the major axes of trait divergence in three independent species radiations.

Results

Machine learning models were able t

SUBMITTER: Majumder S 

PROVIDER: S-EPMC11443442 | biostudies-literature | 2024 Sep-Oct

REPOSITORIES: biostudies-literature