| Abstract Detail
Ecophysiology Majumder, Sambadi [1]. Machine Learning approaches to investigating trait divergence under species diversification . As clades of species diversify and differentiate from one another, a large number of plant traits may change simultaneously. In many cases, this may hinder identification of which traits are most strongly differentiated among species, impairing the formulation of evolutionary hypotheses. In this study we utilize machine learning-based modelling approaches, specifically feature selection, to identify the most strongly differentiated traits among twenty-eight species forming the diploid backbone of the genus Helianthus, the wild sunflowers. We extend this study to further examine the most prominently diverged traits within three major clades within this genus to identify highly differentiated traits in clades differing in life history and that occupy different geographic regions. This was done to examine whether the same traits are consistently important, or if clades are diverging along different axes of multivariate trait space. We further use machine learning-based modelling to facilitate trait-based species classification using the most diverged continuous traits amongst the 28 wild species from a genus and a clade-level perspective, generating dichotomous keys from trait data.
Log in to add this item to your schedule
1 - University Of Central Florida, Department Of Biology, 4110 Libra Drive, Orlando, FL, 32816, United States
Keywords:
Machine learning
sunflower
species concepts
ecophysiology.
Presentation Type: Oral Paper
Session: ECOPH1, Ecophysiology I
Location: /
Date: Monday, July 19th, 2021
Time: 11:45 AM(EDT)
Number: ECOPH1008
Abstract ID:817
Candidate for Awards:Physiological Section Physiological Section Li-COR Prize,Physiological Section Best Paper Presentation |
