Convolutional neural network (CNN) can be used to quickly identify crop seed varieties. 1200 seeds of ten soybean varieties were selected, hyperspectral images of both the front and the back of the seeds were collected, and the reflectance of soybean was derived from the hyperspectral images. A total of 9600 images were obtained after data augmentation, and the images were divided into a training set, validation set, and test set with a 3:1:1 ratio. Pretrained models (AlexNet, ResNet18, Xception, InceptionV3, DenseNet201, and NASNetLarge) after fine-tuning were used for transfer training. The optimal CNN model for soybean seed variety identification was selected. Furthermore, the traditional machine learning models for soybean seed variety identification were established by using reflectance as input. The results show that the six models all achieved 91% accuracy in the validation set and achieved accuracy values of 90.6%, 94.5%, 95.4%, 95.6%, 96.8%, and 97.2%, respectively, in the test set. This method is better than the identification of soybean seed varieties based on hyperspectral reflectance. The experimental results support a novel method for identifying soybean seeds rapidly and accurately, and this method also provides a good reference for the identification of other crop seeds.

Previous studies have revealed a significant genetic relationship between phosphorus (P)-efficiency and photosynthesis-related traits in soybean. In this study, we used proteome profiling in combination with expression analysis, biochemical investigations, and leaf ultrastructural analysis to identify the underlying physiological and molecular responses. The expression analysis and ultrastructural analysis showed that the photosynthesis key genes were decreased at transcript levels and the leaf mesophyll and chloroplast were severely damaged after low-P stress. Approximately 55 protein spots showed changes under low-P condition by mass spectrometry, of which 17 were involved in various photosynthetic processes. Further analysis revealed the depression of photosynthesis caused by low-P stress mainly involves the regulation of leaf structure, adenosine triphosphate (ATP) synthesis, absorption and transportation of CO₂, photosynthetic electron transport, production of assimilatory power, and levels of enzymes related to the Calvin cycle. In summary, our findings indicated that the existence of a stringent relationship between P supply and the genomic control of photosynthesis in soybean. As an important strategy to protect soybean photosynthesis, P could maintain the stability of cell structure, up-regulate the enzymes’ activities, recover the process of photosystem II (PSII), and induce the expression of low-P responsive genes and proteins.

Hyperspectral imaging is a nondestructive testing technology that integrates spectroscopy and iconology technologies, which enables us to quickly obtain both internal and external information of objects and identify crop seed varieties. First, the hyperspectral images of ten soybean seed varieties were collected and the reflectance was obtained. Savitzky-Golay smoothing (SG), first derivative (FD), standard normal variate (SNV), fast Fourier transform (FFT), Hilbert transform (HT), and multiplicative scatter correction (MSC) spectral reflectance pretreatment methods were used. Then, the feature wavelengths and feature information of the pretreated spectral reflectance data were extracted using competitive adaptive reweighted sampling (CARS), the successive projections algorithm (SPA), and principal component analysis (PCA). Finally, 5 classifiers, Bayes, support vector machine (SVM), k-nearest neighbor (KNN), ensemble learning (EL), and artificial neural network (ANN), were used to identify seed varieties. The results showed that MSC-CARS-EL had the highest accuracy among the 90 combinations, with training set, test set, and 5-fold cross-validation accuracies of 100%, 100%, and 99.8%, respectively. Moreover, the contribution of spectral pretreatment to discrimination accuracy was higher than those of feature extraction and classifier selection. Pretreatment methods determined the range of the identification accuracy, feature-selective methods and classifiers only changed within this range. The experimental results provide a good reference for the identification of other crop seed varieties.

Class III peroxidases (PODs), commonly known as secretable class III plant peroxidases, are plant-specific enzymes that play critical roles in not only plant growth and development but also the responses to biotic and abiotic stress. In this study, we identified 198 nonredundant POD genes, designated GhPODs, with 180 PODs being predicted to secrete into apoplast. These POD genes were divided into 10 sub-groups based on their phylogenetic relationships. We performed systematic bioinformatic analysis of the POD genes, including analysis of gene structures, phylogenetic relationships, and gene expression profiles. The GhPODs are unevenly distributed on both upland cotton sub-genome A and D chromosomes. Additionally, these genes have undergone 15 segmental and 12 tandem duplication events, indicating that both segmental and tandem duplication contributed to the expansion of the POD gene family in upland cotton. Ka/Ks analysis suggested that most duplicated GhPODs experienced negative selection, with limited functional divergence during the duplication events. High-throughput RNA-seq data indicated that most highly expressed genes might play significant roles in root, stem, leaf, and fiber development. Under K or P deficiency conditions, PODs showed different expression patterns in cotton root and leaf. This study provides useful information for further functional analysis of the POD gene family in upland cotton.

Seeds stored for a prolonged period are subject to aging and a reduction in germination potential (GP), which will negatively affect seed sales. Rare-earth elements have a synergistic effect on the improvement of seed GP. In this study, we examined the effects of neodymium on biochemical components, the antioxidant protective system, and metabolism-related enzymes during germination of naturally and artificially aged seeds of three wheat cultivars. Seed germination indices, biochemical substance contents, and enzyme activities decreased after seed aging. Soaking seeds in a neodymium nitrate solution revived aged wheat seeds at an optimal concentration of 20 µmol/L for 8 h. Soaking in a neodymium nitrate solution increased the GP4 (by 2.25–60.9%), germination index (by 1.69–29.2%), and vigor index (by 3.36–18.7%) of aged seeds. Compared with non-soaked seeds, soaking significantly changed the contents of biochemical substances, and the activities of antioxidant protective enzymes and metabolic enzymes in seedlings were increased. Soaking with neodymium may revive aged seeds by regulating the synthesis of soluble sugars, soluble proteins, chlorophyll, and carotenoids and decomposing malondialdehyde in the germinating seed. Root dehydrogenase and amylase showed different responses to the aging modes. The differential responses of root dehydrogenase and amylase may reflect differences in the resistance of enzymes to long-term mild seed aging and short-term severe environmental aging.