NIRSpredict

NIRSpredict predicts plant phenotypic traits from near-infrared spectroscopy data using deep learning models trained on Arabidopsis thaliana.

Key Features:

• Trait Prediction: Predicts 81 Arabidopsis thaliana phenotypic traits, including classical functional traits and commonly measured chemical compounds, from near-infrared spectroscopy values.
• Database Access: Provides access to the database of traits used for model calibration, which can be filtered, extracted, and visualized in a global context.
• Database Extension: Allows users to submit their own datasets to extend the database and contribute to application development.

Scientific Applications:

• Plant Phenomics: Facilitates the adoption of phenomics in functional and evolutionary ecology by providing a reliable way of characterizing plant populations across geographical ranges.
• Chemical Compound Prediction: Enables prediction of a large variety of commonly measured chemical compounds in plants alongside functional traits.

Methodology:

NIRSpredict uses deep learning models built from near-infrared spectroscopy values to predict plant traits.