Classification of plant growth-promoting bacteria inoculation status and prediction of growth-related traits in tropical maize using hyperspectral image and genomic data

[摘要] Recent technological advances in high-throughput phenotyping have created newopportunities for the prediction of complex traits. In particular, phenomic predictionusing hyperspectral reflectance could capture various signals that affect phenotypesgenomic prediction might not explain. A total of 360 inbred maize (Zea mays L.) lineswith or without plant growth-promoting bacterial inoculation management undernitrogen stress were evaluated using 150 spectral wavelengths ranging from 386to 1,021 nm and 13,826 single-nucleotide polymorphisms. Six prediction modelswere explored to assess the predictive ability of hyperspectral and genomic data forinoculation status and plant growth-related traits. The best models for hyperspectralprediction were partial least squares and automated machine learning. The Bayesianridge regression and BayesB were the best performers for genomic prediction. Overall, hyperspectral prediction showed greater predictive ability for shoot dry mass andstalk diameter, whereas genomic prediction was better for plant height. The prediction models that simultaneously accommodated both hyperspectral and genomic dataresulted in a predictive ability as high as that of phenomics or genomics alone. Ourresults highlight the usefulness of hyperspectral-based phenotyping for managementand phenomic prediction studies.

[学科分类] 农业科学（综合）