Exploring Soybean Flower and Pod Variation Patterns During Reproductive Period Based on Fusion Deep Learning.

The soybean flower and the pod drop are important factors in soybean yield, and the use of computer vision techniques to obtain the phenotypes of flowers and pods in bulk, as well as in a quick and accurate manner, is a key aspect of the study of the soybean flower and pod drop rate (PDR). This paper compared a variety of deep learning algorithms for identifying and counting soybean flowers and pods, and found that the Faster R-CNN model had the best performance. Furthermore, the Faster R-CNN model was further improved and optimized based on the characteristics of soybean flowers and pods. The accuracy of the final model for identifying flowers and pods was increased to 94.36 and 91%, respectively. Afterward, a fusion model for soybean flower and pod recognition and counting was proposed based on the Faster R-CNN model, where the coefficient of determinationR2 between counts of soybean flowers and pods by the fusion model and manual counts reached 0.965 and 0.98, respectively. The above results show that the fusion model is a robust recognition and counting algorithm that can reduce labor intensity and improve efficiency. Its application will greatly facilitate the study of the variable patterns of soybean flowers and pods during the reproductive period. Finally, based on the fusion model, we explored the variable patterns of soybean flowers and pods during the reproductive period, the spatial distribution patterns of soybean flowers and pods, and soybean flower and pod drop patterns.

Automatic and Accurate Acquisition of Stem-Related Phenotypes of Mature Soybean Based on Deep Learning and Directed Search Algorithms.

The stem-related phenotype of mature stage soybean is important in soybean material selection. How to improve on traditional manual methods and obtain the stem-related phenotype of soybean more quickly and accurately is a problem faced by producers. With the development of smart agriculture, many scientists have explored soybean phenotypes and proposed new acquisition methods, but soybean mature stem-related phenotype studies are relatively scarce. In this study, we used a deep learning method within the convolutional neural network to detect mature soybean stem nodes and identified soybean structural features through a novel directed search algorithm. We subsequently obtained the pitch number, internodal length, branch number, branching angle, plant type spatial conformation, plant height, main stem length, and new phenotype-stem curvature. After 300 epochs, we compared the recognition results of various detection algorithms to select the best. Among them, YOLOX had a maximum average accuracy (mAP) of 94.36% for soybean stem nodes and scale markers. Through comparison of the phenotypic information extracted by the directed search algorithm with the manual measurement results, we obtained the Pearson correlation coefficients, R, of plant height, pitch number, internodal length, main stem length, stem curvature, and branching angle, which were 0.9904, 0.9853, 0.9861, 0.9925, 0.9084, and 0.9391, respectively. These results show that our algorithm can be used for robust measurements and counting of soybean phenotype information, which can reduce labor intensity, improve efficiency, and accelerate soybean breeding.

Identification of Finely Mapped Quantitative Trait Locus and Candidate Gene Mining for the Three-Seeded Pod Trait in Soybean.

The three-seeded pod number is an important trait that positively influences soybean yield. Soybean variety with increased three-seeded pod number contributes to the seed number/plant and higher yield. The candidate genes of the three-seeded pod may be the key for improving soybean yield. In this study, identification and validation of candidate genes for three-seeded pod has been carried out. First, a total of 36 quantitative trait locus (QTL) were detected from the investigation of recombinant inbred lines including 147 individuals derived from a cross between Charleston and Dongning 594 cultivars. Five consensus QTLs were integrated. Second, an introgressed line CSSL-182 carrying the target segment for the trait from the donor parent was selected to verify the consensus QTL based on its phenotype. Third, a secondary group was constructed by backcrossing with CSSL-182, and two QTLs were confirmed. There were a total of 162 genes in the two QTLs. The mining of candidate genes resulted in the annotation of eight genes with functions related to pod and seed sets. Finally, haplotype analysis and quantitative reverse transcriptase real-time PCR were carried to verify the candidate genes. Four of these genes had different haplotypes in the resource group, and the differences in the phenotype were highly significant. Moreover, the differences in the expression of the four genes during pod and seed development were also significant. These four genes were probably related to the development process underlying the three-seeded pod in soybean. Herein, we discuss the past and present studies related to the three-seeded pod trait in soybean.

Automatic and Accurate Calculation of Rice Seed Setting Rate Based on Image Segmentation and Deep Learning.

The rice seed setting rate (RSSR) is an important component in calculating rice yields and a key phenotype for its genetic analysis. Automatic calculations of RSSR through computer vision technology have great significance for rice yield predictions. The basic premise for calculating RSSR is having an accurate and high throughput identification of rice grains. In this study, we propose a method based on image segmentation and deep learning to automatically identify rice grains and calculate RSSR. By collecting information on the rice panicle, our proposed image automatic segmentation method can detect the full grain and empty grain, after which the RSSR can be calculated by our proposed rice seed setting rate optimization algorithm (RSSROA). Finally, the proposed method was used to predict the RSSR during which process, the average identification accuracy reached 99.43%. This method has therefore been proven as an effective, non-invasive method for high throughput identification and calculation of RSSR. It is also applicable to soybean yields, as well as wheat and other crops with similar characteristics.

Responses of Soybean Genes in the Substituted Segments of Segment Substitution Lines Following a Xanthomonas Infection.

Bacterial blight, which is one of the most common soybean diseases, is responsible for considerable yield losses. In this study, a novel Xanthomonas vasicola strain was isolated from the leaves of soybean plants infected with bacterial blight under field conditions. Sequencing the X. vasicola genome revealed type-III effector-coding genes. Moreover, the hrpG deletion mutant was constructed. To identify the soybean genes responsive to HrpG, two chromosome segment substitution lines (CSSLs) carrying the wild soybean genome, but with opposite phenotypes following Xanthomonas inoculations, were used to analyze gene expression networks based on RNA sequencing at three time points after inoculations with wild-type Xanthomonas or the hrpG deletion mutant. To further identify the hub genes underlying soybean responses to HrpG, the genes located on the substituted chromosome segments were examined. Finally, a combined analysis with the QTLs for resistance to Xanthomonas identified 35 hub genes in the substituted chromosomal segments that may help regulate soybean responses to Xanthomonas and HrpG. Furthermore, two candidate genes in the CSSLs might play pivotal roles in response to Xanthomonas.

Analysing the phenotype development of soybean plants using low-cost 3D reconstruction.

With the development of digital agriculture, 3D reconstruction technology has been widely used to analyse crop phenotypes. To date, most research on 3D reconstruction of field crops has been limited to analysis of population characteristics. Therefore, in this study, we propose a method based on low-cost 3D reconstruction technology to analyse the phenotype development during the whole growth period. Based on the phenotypic parameters extracted from the 3D reconstruction model, we identified the "phenotypic fingerprint" of the relevant phenotypes throughout the whole growth period of soybean plants and completed analysis of the plant growth patterns using a logistic growth model. The phenotypic fingerprint showed that, before the R3 period, the growth of the five varieties was similar. After the R5 period, the differences among the five cultivars gradually increased. This result indicates that the phenotypic fingerprint can accurately reveal the patterns of phenotypic changes. The logistic growth model of soybean plants revealed the time points of maximum growth rate of the five soybean varieties, and this information can provide a basis for developing guidelines for water and fertiliser application to crops. These findings will provide effective guidance for breeding and field management of soybean and other crops.

RNA Sequencing-Associated Study Identifies GmDRR1 as Positively Regulating the Establishment of Symbiosis in Soybean.

In soybean (Glycine max)-rhizobium interactions, the type III secretion system (T3SS) of rhizobium plays a key role in regulating host specificity. However, the lack of information on the role of T3SS in signaling networks limits our understanding of symbiosis. Here, we conducted an RNA sequencing analysis of three soybean chromosome segment substituted lines, one female parent and two derived lines with different chromosome-substituted segments of wild soybean and opposite nodulation patterns. By analyzing chromosome-linked differentially expressed genes in the substituted segments and quantitative trait loci (QTL)-assisted selection in the substituted-segment region, genes that may respond to type III effectors to mediate plant immunity-related signaling were identified. To narrow down the number of candidate genes, QTL assistant was used to identify the candidate region consistent with the substituted segments. Furthermore, one candidate gene, GmDRR1, was identified in the substituted segment. To investigate the role of GmDRR1 in symbiosis establishment, GmDRR1-overexpression and RNA interference soybean lines were constructed. The nodule number increased in the former compared with wild-type soybean. Additionally, the T3SS-regulated effectors appeared to interact with the GmDDR1 signaling pathway. This finding will allow the detection of T3SS-regulated effectors involved in legume-rhizobium interactions.

Combined Linkage Mapping and BSA to Identify QTL and Candidate Genes for Plant Height and the Number of Nodes on the Main Stem in Soybean.

Soybean is one of the most important food and oil crops in the world. Plant height (PH) and the number of nodes on the main stem (NNMS) are quantitative traits closely related to soybean yield. In this study, we used 208 chromosome segment substitution lines (CSSL) populations constructed using "SN14" and "ZYD00006" for quantitative trait locus (QTL) mapping of PH and NNMS. Combined with bulked segregant analysis (BSA) by extreme materials, 8 consistent QTLs were identified. According to the gene annotation of the QTL interval, a total of 335 genes were obtained. Five of which were associated with PH and NNMS, potentially representing candidate genes. RT-qPCR of these 5 candidate genes revealed two genes with differential relative expression levels in the stems of different materials. Haplotype analysis showed that different single nucleotide polymorphisms (SNPs) between the excellent haplotypes in Glyma.04G251900 and Glyma.16G156700 may be the cause of changes in these traits. These results provide the basis for research on candidate genes and marker-assisted selection (MAS) in soybean breeding.

Analysis of miRNAs Targeted Storage Regulatory Genes during Soybean Seed Development Based on Transcriptome Sequencing.

Soybeans are an important cash crop and are widely used as a source of vegetable protein and edible oil. MicroRNAs (miRNA) are endogenous small RNA that play an important regulatory role in the evolutionarily conserved system of gene expression. In this study, we selected four lines with extreme phenotypes, as well as high or low protein and oil content, from the chromosome segment substitution line (CSSL) constructed from suinong (SN14) and ZYD00006, and planted and sampled at three stages of grain development for small RNA sequencing and expression analysis. The sequencing results revealed the expression pattern of miRNA in the materials, and predicted miRNA-targeted regulatory genes, including 1967 pairs of corresponding relationships between known-miRNA and their target genes, as well as 597 pairs of corresponding relationships between novel-miRNA and their target genes. After screening and annotating genes that were targeted for regulation, five specific genes were identified to be differentially expressed during seed development and subsequently analyzed for their regulatory relationship with miRNAs. The expression pattern of the targeted gene was verified by Real-time Quantitative PCR (RT-qPCR). Our research provides more information about the miRNA regulatory network in soybeans and further identifies useful genes that regulate storage during soy grain development, providing a theoretical basis for the regulation of soybean quality traits.

QTL analysis of nodule traits and the identification of loci interacting with the type III secretion system in soybean.

Symbiotic nitrogen fixation is the main source of nitrogen for soybean growth. Since the genotypes of rhizobia and soybean germplasms vary, the nitrogen-fixing ability of soybean after inoculation also varies. A few studies have reported that quantitative trait loci (QTLs) control biological nitrogen fixation traits, even soybean which is an important crop. The present study reported that the Sinorhizobium fredii HH103 gene rhcJ belongs to the tts (type III secretion) cluster and that the mutant HH103ΩrhcJ can clearly decrease the number of nodules in American soybeans. However, few QTLs of nodule traits have been identified. This study used a soybean (Glycine max (L.) Merr.) 'Charleston' × 'Dongnong 594' (C × D, n = 150) recombinant inbred line (RIL). Nodule traits were analysed in the RIL population after inoculation with S. fredii HH103 and the mutant HH103ΩrhcJ. Plants were grown in a greenhouse with a 16-h light cycle at 26 °C and an 8-h dark cycle at 18 °C. Then, 4 weeks after inoculation, plants were harvested for evaluation of nodule traits. Through QTL mapping, 16 QTLs were detected on 8 chromosomes. Quantitative PCR (qRT-PCR) and RNA-seq analysis determined that the genes Glyma.04g060600, Glyma.18g159800 and Glyma.13g252600 might interact with rhcJ.

Identification of Soybean Genes Related to Soybean Seed Protein Content Based on Quantitative Trait Loci Collinearity Analysis.

Increasing the protein content of soybean seeds through a higher ratio of glycinin is important for soybean breeding and food processing; therefore, the integration of different quantitative trait loci (QTLs) is of great significance. In this study, we investigated the collinearity of seed protein QTLs. We identified 192 collinear protein QTLs that formed six hotspot regions. The two most important regions had seed protein 36-10 and seed protein 36-20 as hub nodes. We used a chromosome segment substitution line (CSSL) population for QTL validation and identified six CSSL materials with collinear QTLs. Five materials with higher protein and glycinin contents in comparison to the recurrent parent were analyzed. A total of 13 candidate genes related to seed protein from the QTL hotspot intervals were detected, 8 of which had high expression in mature soybean seeds. These results offer a new analysis method for molecular-assisted selection (MAS) and improvement of soybean product quality.

Identification of Soybean Genes Whose Expression is Affected by the Ensifer fredii HH103 Effector Protein NopP.

In some legume⁻rhizobium symbioses, host specificity is influenced by rhizobial nodulation outer proteins (Nops). However, the genes encoding host proteins that interact with Nops remain unknown. We generated an Ensifer fredii HH103 NopP mutant (HH103ΩNopP), and analyzed the nodule number (NN) and nodule dry weight (NDW) of 10 soybean germplasms inoculated with the wild-type E. fredii HH103 or the mutant strain. An analysis of recombinant inbred lines (RILs) revealed the quantitative trait loci (QTLs) associated with NopP interactions. A soybean genomic region containing two overlapping QTLs was analyzed in greater detail. A transcriptome analysis and qRT-PCR assay were used to identify candidate genes encoding proteins that interact with NopP. In some germplasms, NopP positively and negatively affected the NN and NDW, while NopP had different effects on NN and NDW in other germplasms. The QTL region in chromosome 12 was further analyzed. The expression patterns of candidate genes Glyma.12g031200 and Glyma.12g073000 were determined by qRT-PCR, and were confirmed to be influenced by NopP.

Identification of Major QTLs Associated With First Pod Height and Candidate Gene Mining in Soybean.

First pod height (FPH) is a quantitative trait in soybean [Glycine max (L.) Merr.] that affects mechanized harvesting. A compatible combination of the FPH and the mechanized harvester is required to ensure that the soybean is efficiently harvested. In this study, 147 recombinant inbred lines, which were derived from a cross between 'Dongnong594' and 'Charleston' over 8 years, were used to identify the major quantitative trait loci (QTLs) associated with FPH. Using a composite interval mapping method with WinQTLCart (version 2.5), 11 major QTLs were identified. They were distributed on five soybean chromosomes, and 90 pairs of QTLs showed significant epistatic associates with FPH. Of these, 3 were main QTL × main QTL interactions, and 12 were main QTL × non-main QTL interactions. A KEGG gene annotation of the 11 major QTL intervals revealed 8 candidate genes related to plant growth, appearing in the pathways K14486 (auxin response factor 9), K14498 (serine/threonine-protein kinase), and K13946 (transmembrane amino acid transporter family protein), and 7 candidate genes had high expression levels in the soybean stems. These results will aid in building a foundation for the fine mapping of the QTLs related to FPH and marker-assisted selection for breeding in soybean.

Meta-analysis and transcriptome profiling reveal hub genes for soybean seed storage composition during seed development.

Soybean is an important crop providing edible oil and protein source. Soybean oil and protein contents are quantitatively inherited and significantly affected by environmental factors. In this study, meta-analysis was conducted based on soybean physical maps to integrate quantitative trait loci (QTLs) from multiple experiments in different environments. Meta-QTLs for seed oil, fatty acid composition, and protein were identified. Of them, 11 meta-QTLs were located on hot regions for both seed oil and protein. Next, we selected 4 chromosome segment substitution lines with different seed oil and protein contents to characterize their 3 years of phenotype selection in the field. Using strand-specific RNA-sequencing analysis, we profile the time-course transcriptome patterns of soybean seeds at early maturity, middle maturity, and dry seed stages. Pairwise comparison and K-means clustering analysis revealed 7,482 differentially expressed genes and 45 expression patterns clusters. Weighted gene coexpression network analysis uncovered 46 modules of gene expression patterns. The 2 most significant coexpression networks were visualized, and 7 hub genes were identified that were involved in soybean oil and seed storage protein accumulation processes. Our results provided a transcriptome dataset for soybean seed development, and the candidate hub genes represent a foundation for further research.

Discovering Numerical Differences between Animal and Plant microRNAs.

Previous studies have confirmed that there are many differences between animal and plant microRNAs (miRNAs), and that numerical features based on sequence and structure can be used to predict the function of individual miRNAs. However, there is little research regarding numerical differences between animal and plant miRNAs, and whether a single numerical feature or combination of features could be used to distinguish animal and plant miRNAs or not. Therefore, in current study we aimed to discover numerical features that could be used to accomplish this. We performed a large-scale analysis of 132 miRNA numerical features, and identified 17 highly significant distinguishing features. However, none of the features independently could clearly differentiate animal and plant miRNAs. By further analysis, we found a four-feature subset that included helix number, stack number, length of pre-miRNA, and minimum free energy, and developed a logistic classifier that could distinguish animal and plant miRNAs effectively. The precision of the classifier was greater than 80%. Using this tool, we confirmed that there were universal differences between animal and plant miRNAs, and that a single feature was unable to adequately distinguish the difference. This feature set and classifier represent a valuable tool for identifying differences between animal and plant miRNAs at a molecular level.

SNP-SNP Interaction Analysis on Soybean Oil Content under Multi-Environments.

Soybean oil content is one of main quality traits. In this study, we used the multifactor dimensionality reduction (MDR) method and a soybean high-density genetic map including 5,308 markers to identify stable single nucleotide polymorphism (SNP)-SNP interactions controlling oil content in soybean across 23 environments. In total, 36,442,756 SNP-SNP interaction pairs were detected, 1865 of all interaction pairs associated with soybean oil content were identified under multiple environments by the Bonferroni correction with p <3.55×10-11. Two and 1863 SNP-SNP interaction pairs detected stable across 12 and 11 environments, respectively, which account around 50% of total environments. Epistasis values and contribution rates of stable interaction (the SNP interaction pairs were detected in more than 2 environments) pairs were detected by the two way ANOVA test, the available interaction pairs were ranged 0.01 to 0.89 and from 0.01 to 0.85, respectively. Some of one side of the interaction pairs were identified with previously research as a major QTL without epistasis effects. The results of this study provide insights into the genetic architecture of soybean oil content and can serve as a basis for marker-assisted selection breeding.

QTL Location and Epistatic Effect Analysis of 100-Seed Weight Using Wild Soybean (Glycine soja Sieb. & Zucc.) Chromosome Segment Substitution Lines.

Increasing the yield of soybean (Glycine max L. Merrill) is a main aim of soybean breeding. The 100-seed weight is a critical factor for soybean yield. To facilitate genetic analysis of quantitative traits and to improve the accuracy of marker-assisted breeding in soybean, a valuable mapping population consisting of 194 chromosome segment substitution lines (CSSLs) was developed. In these lines, different chromosomal segments of the Chinese cultivar Suinong 14 were substituted into the genetic background of wild soybean (Glycine soja Sieb. & Zucc.) ZYD00006. Based on these CSSLs, a genetic map covering the full genome was generated using 121 simple sequence repeat (SSR) markers. In the quantitative trait loci (QTL) analysis, twelve main effect QTLs (qSW-B1-1/2/3, qSW-D1b-1/2, qSW-D2-1/2, qSW-G-1/2/3, qSW-M-2 and qSW-N-2) underlying 100-seed weight were identified in 2011 and 2012. The epistatic effects of pairwise interactions between markers were analyzed in 2011 and 2012. The results clearly demonstrated that these CSSLs could be used to identify QTLs, and that an epistatic analysis was able to detect several sites with important epistatic effects on 100-seed weight. Thus, we identified loci that will be valuable for improving soybean 100-seed weight. These results provide a valuable foundation for identifying the precise location of genes of interest, and for designing cloning and marker-assisted selection breeding strategies targeting the 100-seed weight of soybean.