High-throughput phenotyping models for quality traits in peanut kernels

Abstract

Peanut is one of the important oil crops. Its kernel quality directly affects its processing characteristics and is an important index for peanut quality evaluation. Establishing a high-throughput phenotyping model for peanut kernel quality and evaluating peanut kernel quality quickly and efficiently might significantly improve the efficiency of peanut breeding. In this study, the spectra of 175 peanut kernel samples (140 RIL populations derived from Yuhua 14 × LOP 215 and 35 other breeding lines) were collected by Antaris II Fourier Transform Near Infrared Spectroscopy Analyzer (Thermo company), and the oil content, protein content, sugar content, and fatty acid content of seed kernel were determined by Soxhlet extraction method, Dumas nitrogen method, anthrone colorimetry, and gas chromatography, respectively. Partial least squares (PLS) was used to construct the near-infrared calibration models of oil content, protein content, sugar content and some fatty acid content of peanut kernel. 30 other peanut materials which were not involved in the modelling were selected to verify the model externally. The determination coefficients (R2) of the models were greater than 0.90, indicating that the models could be applied to the high-throughput prediction of peanut kernel quality traits. This study provides a detection platform for high-throughput phenotypic analysis of peanut kernel quality traits.

Plant phenotyping relevance

近赤外分光とPLSによるピーナッツ種子品質形質の高スループット推定モデルを構築し、外部検証しており、形質取得・予測法が研究の中心である。

Code and data availability

公開論文であることは確認できましたが、現在の公式API・許可済み取得経路では本文を自動取得できませんでした。

No evidence-backed public reproduction asset is currently recorded.