Using Machine Learning Methods Combined with Vegetation Indices and Growth Indicators to Predict Seed Yield of Bromus inermis.

Smooth bromegrass (Bromus inermis) is a perennial, high-quality forage grass. However, its seed yield is influenced by agronomic practices, climatic conditions, and the growing year. The rapid and effective prediction of seed yield can assist growers in making informed production decisions and reducing agricultural risks. Our field trial design followed a completely randomized block design with four blocks and three nitrogen levels (0, 100, and 200 kg·N·ha-1) during 2022 and 2023. Data on the remote vegetation index (RVI), the normalized difference vegetation index (NDVI), the leaf nitrogen content (LNC), and the leaf area index (LAI) were collected at heading, anthesis, and milk stages. Multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) regression models were utilized to predict seed yield. In 2022, the results indicated that nitrogen application provided a sufficiently large range of variation of seed yield (ranging from 45.79 to 379.45 kg ha⁻¹). Correlation analysis showed that the indices of the RVI, the NDVI, the LNC, and the LAI in 2022 presented significant positive correlation with seed yield, and the highest correlation coefficient was observed at the heading stage. The data from 2022 were utilized to formulate a predictive model for seed yield. The results suggested that utilizing data from the heading stage produced the best prediction performance. SVM and RF outperformed MLR in prediction, with RF demonstrating the highest performance (R2 = 0.75, RMSE = 51.93 kg ha-1, MAE = 29.43 kg ha-1, and MAPE = 0.17). Notably, the accuracy of predicting seed yield for the year 2023 using this model had decreased. Feature importance analysis of the RF model revealed that LNC was a crucial indicator for predicting smooth bromegrass seed yield. Further studies with an expanded dataset and integration of weather data are needed to improve the accuracy and generalizability of the model and adaptability for the growing year.

Genomic Identification and Expression Profiling of Lesion Simulating Disease Genes in Alfalfa (Medicago sativa) Elucidate Their Responsiveness to Seed Vigor.

Seed aging, a common physiological phenomenon during forage seed storage, is a crucial factor contributing to a loss of vigor, resulting in delayed seed germination and seedling growth, as well as limiting the production of hay. Extensive bodies of research are dedicated to the study of seed aging, with a particular focus on the role of the production and accumulation of reactive oxygen species (ROS) and the ensuing oxidative damage during storage as a primary cause of decreases in seed vigor. To preserve optimal seed vigor, ROS levels must be regulated. The excessive accumulation of ROS can trigger programmed cell death (PCD), which causes the seed to lose vigor permanently. LESION SIMULATING DISEASE (LSD) is one of the proteins that regulate PCD, encodes a small C2C2 zinc finger protein, and plays a molecular function as a transcriptional regulator and scaffold protein. However, genome-wide analysis of LSD genes has not been performed for alfalfa (Medicago sativa), as one of the most important crop species, and, presently, the molecular regulation mechanism of seed aging is not clear enough. Numerous studies have also been unable to explain the essence of seed aging for LSD gene regulating PCD and affecting seed vigor. In this study, we obtained six MsLSD genes in total from the alfalfa (cultivar Zhongmu No. 1) genome. Phylogenetic analysis demonstrated that the MsLSD genes could be classified into three subgroups. In addition, six MsLSD genes were unevenly mapped on three chromosomes in alfalfa. Gene duplication analysis demonstrated that segmental duplication was the key driving force for the expansion of this gene family during evolution. Expression analysis of six MsLSD genes in various tissues and germinating seeds presented their different expressions. RT-qPCR analysis revealed that the expression of three MsLSD genes, including MsLSD2, MsLSD5, and MsLSD6, was significantly induced by seed aging treatment, suggesting that they might play an important role in maintaining seed vigor. Although this finding will provide valuable insights into unveiling the molecular mechanism involved in losing vigor and new strategies to improve alfalfa seed germinability, additional research must comprehensively elucidate the precise pathways through which the MsLSD genes regulate seed vigor.

Genome-wide analysis and expression profiling of glyoxalase gene families in oat (Avena sativa) indicate their responses to abiotic stress during seed germination.

Abiotic stresses have deleterious effects on seed germination and seedling establishment, leading to significant crop yield losses. Adverse environmental conditions can cause the accumulation of methylglyoxal (MG) within plant cells, which can negatively impact plant growth and development. The glyoxalase system, which consists of the glutathione (GSH)-dependent enzymes glyoxalase I (GLX1) and glyoxalase II (GLX2), as well as the GSH-independent glyoxalase III (GLX3 or DJ-1), plays a crucial role in detoxifying MG. However, genome-wide analysis of glyoxalase genes has not been performed for one of the agricultural important species, oat (Avena sativa). This study identified a total of 26 AsGLX1 genes, including 8 genes encoding Ni2+-dependent GLX1s and 2 genes encoding Zn2+-dependent GLX1s. Additionally, 14 AsGLX2 genes were identified, of which 3 genes encoded proteins with both lactamase B and hydroxyacylglutathione hydrolase C-terminal domains and potential catalytic activity, and 15 AsGLX3 genes encoding proteins containing double DJ-1 domains. The domain architecture of the three gene families strongly correlates with the clades observed in the phylogenetic trees. The AsGLX1, AsGLX2, and AsGLX3 genes were evenly distributed in the A, C, and D subgenomes, and gene duplication of AsGLX1 and AsGLX3 genes resulted from tandem duplications. Besides the core cis-elements, hormone responsive elements dominated the promoter regions of the glyoxalase genes, and stress responsive elements were also frequently observed. The subcellular localization of glyoxalases was predicted to be primarily in the cytoplasm, chloroplasts, and mitochondria, with a few presents in the nucleus, which is consistent with their tissue-specific expression. The highest expression levels were observed in leaves and seeds, indicating that these genes may play important roles in maintaining leaf function and ensuring seed vigor. Moreover, based on in silico predication and expression pattern analysis, AsGLX1-7A, AsGLX2-5D, AsDJ-1-5D, AsGLX1-3D2, and AsGLX1-2A were suggested as promising candidate genes for improving stress resistance or seed vigor in oat. Overall, the identification and analysis of the glyoxalase gene families in this study can provide new strategies for improving oat stress resistance and seed vigor.

Integrating optical imaging techniques for a novel approach to evaluate Siberian wild rye seed maturity.

Advances in optical imaging technology using rapid and non-destructive methods have led to improvements in the efficiency of seed quality detection. Accurately timing the harvest is crucial for maximizing the yield of higher-quality Siberian wild rye seeds by minimizing excessive shattering during harvesting. This research applied integrated optical imaging techniques and machine learning algorithms to develop different models for classifying Siberian wild rye seeds based on different maturity stages and grain positions. The multi-source fusion of morphological, multispectral, and autofluorescence data provided more comprehensive information but also increases the performance requirements of the equipment. Therefore, we employed three filtering algorithms, namely minimal joint mutual information maximization (JMIM), information gain, and Gini impurity, and set up two control methods (feature union and no-filtering) to assess the impact of retaining only 20% of the features on the model performance. Both JMIM and information gain revealed autofluorescence and morphological features (CIELab A, CIELab B, hue and saturation), with these two filtering algorithms showing shorter run times. Furthermore, a strong correlation was observed between shoot length and morphological and autofluorescence spectral features. Machine learning models based on linear discriminant analysis (LDA), random forests (RF) and support vector machines (SVM) showed high performance (>0.78 accuracies) in classifying seeds at different maturity stages. Furthermore, it was found that there was considerable variation in the different grain positions at the maturity stage, and the K-means approach was used to improve the model performance by 5.8%-9.24%. In conclusion, our study demonstrated that feature filtering algorithms combined with machine learning algorithms offer high performance and low cost in identifying seed maturity stages and that the application of k-means techniques for inconsistent maturity improves classification accuracy. Therefore, this technique could be employed classification of seed maturity and superior physiological quality for Siberian wild rye seeds.

Single Seed Identification in Three Medicago Species via Multispectral Imaging Combined with Stacking Ensemble Learning.

Multispectral imaging (MSI) has become a new fast and non-destructive detection method in seed identification. Previous research has usually focused on single models in MSI data analysis, which always employed all features and increased the risk to efficiency and that of system cost. In this study, we developed a stacking ensemble learning (SEL) model for successfully identifying a single seed of sickle alfalfa (Medicago falcata), hybrid alfalfa (M. varia), and alfalfa (M. sativa). SEL adopted a three-layer structure, i.e., level 0 with principal component analysis (PCA), linear discriminant analysis (LDA), and quadratic discriminant analysis (QDA) as models of dimensionality reduction and feature extraction (DRFE); level 1 with support vector machine (SVM), multiple logistic regression (MLR), generalized linear models with elastic net regularization (GLMNET), and eXtreme Gradient Boosting (XGBoost) as basic learners; and level 3 with XGBoost as meta-learner. We confirmed that the values of overall accuracy, kappa, precision, sensitivity, specificity, and sensitivity in the SEL model were all significantly higher than those in basic models alone, based on both spectral features and a combination of morphological and spectral features. Furthermore, we also developed a feature filtering process and successfully selected 5 optimal features out of 33 ones, which corresponded to the contents of chlorophyll, anthocyanin, fat, and moisture in seeds. Our SEL model in MSI data analysis provided a new way for seed identification, and the feature filter process potentially could be used widely for development of a low-cost and narrow-channel sensor.

Genome-Wide Analysis and Expression Profiling of Glutathione Reductase Gene Family in Oat (Avena sativa) Indicate Their Responses to Abiotic Stress during Seed Imbibition.

Abiotic stress disturbs plant cellular redox homeostasis, inhibiting seed germination and plant growth. This is a crucial limitation to crop yield. Glutathione reductase (GR) is an important component of the ascorbate-glutathione (AsA-GSH) cycle which is involved in multiple plant metabolic processes. In the present study, GRs in A. sativa (AsGRs) were selected to explore their molecular characterization, phylogenetic relationship, and RNA expression changes during seed imbibition under abiotic stress. Seven AsGR genes were identified and mapped on six chromosomes of A, C, and D subgenomes. Phylogenetic analysis and subcellular localization of AsGR proteins divided them into two sub-families, AsGR1 and AsGR2, which were predicted to be mainly located in cytoplasm, mitochondrion, and chloroplast. Cis-elements relevant to stress and hormone responses are distributed in promoter regions of AsGRs. Tissue-specific expression profiling showed that AsGR1 genes were highly expressed in roots, leaves, and seeds, while AsGR2 genes were highly expressed in leaves and seeds. Both AsGR1 and AsGR2 genes showed a decreasing-increasing expression trend during seed germination under non-stress conditions. In addition, their responses to drought, salt, cold, copper, H2O2, and ageing treatments were quite different during seed imbibition. Among the seven AsGR genes, AsGR1-A, AsGR1-C, AsGR2-A, and AsGR2-D responded more significantly, especially under drought, ageing, and H2O2 stress. This study has laid the ground for the functional characterization of GR and the improvement of oat stress tolerance and seed vigor.

Dynamic Responses of Antioxidant and Glyoxalase Systems to Seed Aging Based on Full-Length Transcriptome in Oat (Avena sativa L.).

Seed aging is a major challenge for food security, agronomic production, and germplasm conservation, and reactive oxygen species (ROS) and methylglyoxal (MG) are highly involved in the aging process. However, the regulatory mechanisms controlling the abundance of ROS and MG are not well characterized. To characterize dynamic response of antioxidant and glyoxalase systems during seed aging, oat (Avena sativa L.) aged seeds with a range of germination percentages were used to explore physiological parameters, biochemical parameters and relevant gene expression. A reference transcriptome based on PacBio sequencing generated 67,184 non-redundant full-length transcripts, with 59,050 annotated. Subsequently, eleven seed samples were used to investigate the dynamic response of respiration, ROS and MG accumulation, antioxidant enzymes and glyoxalase activity, and associated genes expression. The 48 indicators with high correlation coefficients were divided into six major response patterns, and were used for placing eleven seed samples into four groups, i.e., non-aged (Group N), higher vigor (Group H), medium vigor (Group M), and lower vigor (Group L). Finally, we proposed a putative model for aging response and self-detoxification mechanisms based on the four groups representing different aging levels. In addition, the outcomes of the study suggested the dysfunction of antioxidant and glyoxalase system, and the accumulation of ROS and MG definitely contribute to oat seed aging.