Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel

Abstract Estimates of plant traits derived from hyperspectral reflectance data have the potential to efficiently substitute for traits, which are time or labor intensive to manually score. Typical workflows for estimating plant traits from hyperspectral reflectance data employ supervised classificat...

Full description

Saved in:
Bibliographic Details
Main Authors: Michael C. Tross, Marcin W. Grzybowski, Talukder Z. Jubery, Ryleigh J. Grove, Aime V. Nishimwe, J. Vladimir Torres‐Rodriguez, Guangchao Sun, Baskar Ganapathysubramanian, Yufeng Ge, James C. Schnable
Format: Article
Language:English
Published: Wiley 2024-12-01
Series:Plant Phenome Journal
Online Access:https://doi.org/10.1002/ppj2.20106
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850105116883943424
author Michael C. Tross
Marcin W. Grzybowski
Talukder Z. Jubery
Ryleigh J. Grove
Aime V. Nishimwe
J. Vladimir Torres‐Rodriguez
Guangchao Sun
Baskar Ganapathysubramanian
Yufeng Ge
James C. Schnable
author_facet Michael C. Tross
Marcin W. Grzybowski
Talukder Z. Jubery
Ryleigh J. Grove
Aime V. Nishimwe
J. Vladimir Torres‐Rodriguez
Guangchao Sun
Baskar Ganapathysubramanian
Yufeng Ge
James C. Schnable
author_sort Michael C. Tross
collection DOAJ
description Abstract Estimates of plant traits derived from hyperspectral reflectance data have the potential to efficiently substitute for traits, which are time or labor intensive to manually score. Typical workflows for estimating plant traits from hyperspectral reflectance data employ supervised classification models that can require substantial ground truth datasets for training. We explore the potential of an unsupervised approach, autoencoders, to extract meaningful traits from plant hyperspectral reflectance data using measurements of the reflectance of 2151 individual wavelengths of light from the leaves of maize (Zea mays) plants harvested from 1658 field plots in a replicated field trial. A subset of autoencoder‐derived variables exhibited significant repeatability, indicating that a substantial proportion of the total variance in these variables was explained by difference between maize genotypes, while other autoencoder variables appear to capture variation resulting from changes in leaf reflectance between different batches of data collection. Several of the repeatable latent variables were significantly correlated with other traits scored from the same maize field experiment, including one autoencoder‐derived latent variable (LV8) that predicted plant chlorophyll content modestly better than a supervised model trained on the same data. In at least one case, genome‐wide association study hits for variation in autoencoder‐derived variables were proximal to genes with known or plausible links to leaf phenotypes expected to alter hyperspectral reflectance. In aggregate, these results suggest that an unsupervised, autoencoder‐based approach can identify meaningful and genetically controlled variation in high‐dimensional, high‐throughput phenotyping data and link identified variables back to known plant traits of interest.
format Article
id doaj-art-2f45c21e70ec4f1aa738ded223ea579c
institution DOAJ
issn 2578-2703
language English
publishDate 2024-12-01
publisher Wiley
record_format Article
series Plant Phenome Journal
spelling doaj-art-2f45c21e70ec4f1aa738ded223ea579c2025-08-20T02:39:11ZengWileyPlant Phenome Journal2578-27032024-12-0171n/an/a10.1002/ppj2.20106Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panelMichael C. Tross0Marcin W. Grzybowski1Talukder Z. Jubery2Ryleigh J. Grove3Aime V. Nishimwe4J. Vladimir Torres‐Rodriguez5Guangchao Sun6Baskar Ganapathysubramanian7Yufeng Ge8James C. Schnable9Quantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USADepartment of Mechanical Engineering Iowa State University Ames Iowa USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USADepartment of Mechanical Engineering Iowa State University Ames Iowa USACenter for Plant Science Innovation University of Nebraska‐Lincoln Lincoln Nebraska USAQuantitative Life Sciences Initiative University of Nebraska‐Lincoln Lincoln Nebraska USAAbstract Estimates of plant traits derived from hyperspectral reflectance data have the potential to efficiently substitute for traits, which are time or labor intensive to manually score. Typical workflows for estimating plant traits from hyperspectral reflectance data employ supervised classification models that can require substantial ground truth datasets for training. We explore the potential of an unsupervised approach, autoencoders, to extract meaningful traits from plant hyperspectral reflectance data using measurements of the reflectance of 2151 individual wavelengths of light from the leaves of maize (Zea mays) plants harvested from 1658 field plots in a replicated field trial. A subset of autoencoder‐derived variables exhibited significant repeatability, indicating that a substantial proportion of the total variance in these variables was explained by difference between maize genotypes, while other autoencoder variables appear to capture variation resulting from changes in leaf reflectance between different batches of data collection. Several of the repeatable latent variables were significantly correlated with other traits scored from the same maize field experiment, including one autoencoder‐derived latent variable (LV8) that predicted plant chlorophyll content modestly better than a supervised model trained on the same data. In at least one case, genome‐wide association study hits for variation in autoencoder‐derived variables were proximal to genes with known or plausible links to leaf phenotypes expected to alter hyperspectral reflectance. In aggregate, these results suggest that an unsupervised, autoencoder‐based approach can identify meaningful and genetically controlled variation in high‐dimensional, high‐throughput phenotyping data and link identified variables back to known plant traits of interest.https://doi.org/10.1002/ppj2.20106
spellingShingle Michael C. Tross
Marcin W. Grzybowski
Talukder Z. Jubery
Ryleigh J. Grove
Aime V. Nishimwe
J. Vladimir Torres‐Rodriguez
Guangchao Sun
Baskar Ganapathysubramanian
Yufeng Ge
James C. Schnable
Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
Plant Phenome Journal
title Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
title_full Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
title_fullStr Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
title_full_unstemmed Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
title_short Data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
title_sort data driven discovery and quantification of hyperspectral leaf reflectance phenotypes across a maize diversity panel
url https://doi.org/10.1002/ppj2.20106
work_keys_str_mv AT michaelctross datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT marcinwgrzybowski datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT talukderzjubery datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT ryleighjgrove datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT aimevnishimwe datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT jvladimirtorresrodriguez datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT guangchaosun datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT baskarganapathysubramanian datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT yufengge datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel
AT jamescschnable datadrivendiscoveryandquantificationofhyperspectralleafreflectancephenotypesacrossamaizediversitypanel