Assessment and phenotypic identification of millet germplasm (Setaria italica L.) in Liaoning, China

TL;DR

This study analyzed millet germplasm in China to identify high-yield and desirable traits for breeding and agriculture.

Contribution

The study introduces a multivariate analysis approach to identify high-yield millet germplasm resources with specific desirable traits.

Findings

Millet germplasm resources showed abundant phenotypic diversity and complex trait correlations.

Four specific millet resources were identified for breeding due to their high yield and desirable traits.

Subear internode length was found to be a useful indicator for maturity selection.

Abstract

This study evaluated millet germplasms in Liaoning Province to support the collection, preservation and innovation of millet germplasm resources.

The study was conducted from 2018 to 2020, involved the selection of 105 millet germplasm resources from the Germplasm Bank of the Liaoning Academy of Agricultural Sciences (LAAS), the observation and recording of 31 traits, and the application of multivariate analysis methods to assess phenotypic diversity.

From the diversity analysis and correlation analysis, it was found that the tested traits had abundant diversity and complex correlations among them. Principal component analysis (PCA) comprehensively analyzed all quantitative traits and extracted seven principal components. Grey relational analysis (GRA) highlighted the varied contributions of different traits to yield. Through systematic cluster analysis (SCA), the resources were categorized into six groups at Euclidean distance of 17.09. K-mean cluster analysis determined the distribution interval and central value of each trait, then identified resources with desirable traits.

The results revealed resources that possess characteristics such as upthrow seedling leaves, more tillers and branches, larger and well-formed ears, and lodging resistance prefer to higher grain yield. It was also discovered that the subear internode length (SIL) could be an indicator for maturity selection. Four specific resources, namely, Dungu No. 1, Xiao-li-xiang, Basen Shengu, and Yuhuanggu No. 1, were identified for further breeding and practical applications.

Figures