High-throughput phenotyping of photosynthetic capacities: an ensemble approach based on multiple machine learning models

Increasing demands for food, fiber, and fuel caused by rising human population and higher living standards may not be satisfied with the world's current farmlands stressed by a changing climate. This conundrum can be alleviated by providing highly photosynthetically efficient crop cultivars to farmers. Current research efforts to increase photosynthetic energy conversion efficiency has produced a wealth of photosynthetic information at genomic and molecular levels, which are yet to be linked to phenotype in a real-world environment efficiently. Though partial least squares regression (PLSR) has been commonly used to relate hyperspectral reflectance to photosynthetic parameters, its modeling performance varies significantly across different plant species, regions, and growth environments. Thus, to cope with the heterogenous performances of PLSR among different situations, this study proposes a new approach to phenotype photosynthetic parameters. Six tobacco genotypes including both transgenic and wild type lines were used to test the developed approach. Leaf reflectance of the six genotypes were measured from 400-2500 nm using a high-spectral-resolution spectroradiometer. The corresponding response of photosynthesis to intercellular CO₂ concentration for each leaf was captured using a portable leaf gas exchange system. The stacked regression was achieved by combining six empirical and machine learning models, including artificial neural network (ANN), support vector machine (SVM), least absolute shrinkage and selection operator (LASSO), random forest, Gaussian process, and PLSR. Results suggested that the stacked regression had a better predication performance than individual regression algorithms in phenotyping of photosynthetic parameters. Analysis of variable importance also revealed common spectral bands used for predicting photosynthetic parameters. Further refinements will be made to understand the portability of the stacked regression to phenotype other biophysical parameters.