Leaf, as a photosynthetic organ of crops, its senescence process has an important impact on yield formation, but the relationship between leaf senescence and phyllosphere microorganisms has been less studied. In order to explore the impact of the senescence process of maize leaves on the phyllosphere bacterial community, this study used three maize varieties of different maturity time (early-maturation variety Heike Yu 17 (H17), mid-maturation variety Zhongdan 111 (Z111), and late-maturation variety Shen Yu 21 (S21) in Northeast China as the experimental materials, and the leaves of the ear position of the three maize varieties were sampled five times starting from the blooming stage of early- maturation varieties, and the physiological indexes of senescence were determined. And at the same time, the community composition of endogenous and exogenous bacteria in/on the leaves was determined based on high-throughput sequencing technology. The results showed that at the late reproductive stage, leaf water content, POD and SOD activities were significantly higher in the mid- and late-maturation varieties than in the early-maturation varieties. At the phylum level, Cyanobacteria were endemic to mid- and late-maturation cultivars; at the genus level, the relative abundance of the endogenous shared bacteria Sphingomonas, Methylobacterium, and Deinococcus in maize leaves decreased significantly at later stages of maturation (IV and V). The relative abundance of endogenous bacteria Streptomyces and exogenous bacteria P3OB-42 were significantly enriched in the late senescence period, with similar trends and significant differences in relative abundance among the three species. The relative abundance of endogenous and exogenous bacteria differed significantly, with the top 5 exogenous bacteria accounting for more than 60%, while for endogenous bacteria, the top 5 accounted for only more than 30%. Soluble sugar content, photosynthetic pigment content and SOD activity were significantly correlated with bacterial community structure and abundance. In conclusion, mid- and late-maturation varieties were effective in prolonging leaf greening period, maintaining late leaf physiological activity with delaying senescence. The effects of senescence on the composition and diversity of endogenous bacterial communities were significantly greater than those of exogenous bacteria, and there were significantly different genera among three maize varieties studied. Moreover, soluble sugar content, photosynthetic pigment content and SOD activity were the key factors affecting the phyllosphere bacterial communities as well as the dominant species.

Multi-environment field testing is an important way to select optimize maize yield and yield stability varieties. However, because of its high cost, it has gradually become a challenge in plant breeding. The combination of field sparse testing and genome-wide prediction method can be used to predict untested phenotypes, reduced the effort and cost on field testing. In this experiment, 244 inbred lines of natural populations were planted in Shunyi, Beijing and Mishan, Heilongjiang in 2022 and 2023. Six agronomic traits were studied, including days to anthesis, plant height, ear height, ear length, kernel number per row and ear row number. The effects of four different models (Single, Across, M×E and R-norm), two different cross-validation schemes (CV1 and CV2) and three different training sets sampling ratios (0.5, 0.7 and 0.9) on the prediction accuracy were compared. The results showed that the average prediction accuracy of the six agronomic traits was 0.67, 0.58, 0.50, 0.33, 0.33 and 0.48. The average prediction accuracy of the Single model, Across model, M×E model and R-norm model was 0.36, 0.52, 0.53 and 0.53 for each trait. In CV1, the average prediction accuracy of each model in six traits ranged from 0.19 to 0.65, and in CV2, the average prediction accuracy ranged from 0.47 to 0.89. The comparison of different training set sampling ratios shows that the improvement of the proportion of the training sets has limited improvement in the prediction accuracy of different traits in different models, and the maximum is only 0.05. The results show that the CV2 training set can be used to form a scheme and include phenotypic data from multiple environments in the prediction model to provide good prediction accuracy for multi-environment prediction.