Kernel Width Research Articles

Robust adaptive generalized correntropy-based smoothed graph signal recovery with a kernel width learning

Robust adaptive generalized correntropy-based smoothed graph signal recovery with a kernel width learning

Identification and validation of two quantitative trait loci showing pleiotropic effect on multiple grain-related traits in bread wheat (Triticum aestivum L.).

QKl/Tgw/Gns.yaas-2D associates with KL, TGW, and GNS, and QKl/Tgw.yaas-5A associates with KL and TGW. Significantly pleiotropic and additive effects of these two QTL were validated. The YM5 allele both at QKl/Tgw/Gns.yaas-2D and QKl/Tgw.yaas-5A was proved to be the best allelic combination for improving yield potential. Kernel length (KL), kernel width (KW), thousand grain weight (TGW), and grain number per spike (GNS) play important roles in the yield improvement of wheat. In this study, one recombinant inbred line (RIL) derived from a cross between Yangmai 5 (YM5) and Yanzhan 1 (YZ1) was used to identify quantitative trait loci (QTL) associated with KL, KW, TGW, and GNS across three years. Two pleiotropic QTL namely QKl/Tgw/Gns.yaas-2D and QKl/Tgw.yaas-5A were located in two genomic regions on chromosomes 2D and 5A, respectively. Breeder-friendly Kompetitive Allele-Specific PCR (KASP) markers for QKl/Tgw/Gns.yaas-2D and QKl/Tgw.yaas-5A were developed and validated in a set of 246 wheat cultivars/lines. Analysis of allelic combinations indicated that the YM5 allele both at QKl/Tgw/Gns.yaas-2D and QKl/Tgw.yaas-5A is probably the best one to promote TGW, GNS, and grain weight per spike. Based on the analysis of gene annotation, sequence variations, expression patterns, and GO enrichment, twenty-five and twenty-four candidate genes of QKl/Tgw/Gns.yaas-2D and QKl/Tgw.yaas-5A, respectively, were identified. These results provide the basis of fine-mapping the target QTL and marker-assisted selection in wheat yield-breeding programs.

Applications of Machine Learning: Classifying Wheat Seeds through Computer Neural Networks

In the past few decades, the incompetent processes of traditional mechanical seed sorting have given rise to new, innovative methods, along with the integration of smart technology in agriculture. The present study created a multi-classification machine learning model that classifies wheat seeds based on given characteristics by giving highly accurate predictions for future samples based on given data. Due to various uses of different types of seeds, effective sorting is an integral preparatory stage for food processing. The dataset used in the study included seven features (directly measured characteristics of seeds), three labels (types of seeds), and around two hundred examples. The seven features include area, perimeter, compactness, kernel length, kernel width, asymmetry coefficient, and kernel groove. The first model was created through neural networks trained on all seven characteristics, reaching an overall accuracy of 0.94. Implementing a decision tree method, feature selection was applied to the data, which distinguished kernel groove, kernel width, and asymmetry coefficient as the three most important features. A new model was constructed with only the three selected features, which reduced the risk of overfitting, eliminated noise, and yielded a higher accuracy (0.96) than the previous model. Both multi-classification models used the softmax activation function and results were compared; feature selection noticeably reduced any inaccuracies, including categorical cross entropy loss. The main incentive of the study was to develop an effective model to sort wheat seeds (to replace traditional mechanical sorting) and provide reasonable recommendations for practical use through an analysis of feature selection.

Persistent biological invasions alter ecological network topology, impacting disease transmission during community assembly

Ecological networks experiencing persistent biological invasions may exhibit distinct topological properties, complicating the understanding of how network topology affects disease transmission during invasion-driven community assembly. We developed a trait-based network model to assess the impact of network topology on disease transmission, measured as community- and species-level disease prevalence. We found that trait-based feeding interactions between host species determine the frequency distribution of the niche of co-occurring species in steady-state communities, being either bimodal or multimodal. The width of the growth kernel influences the degree-biomass relationship of species, being either weakly positive or strongly negative. When this relationship is weakly positive, species-level disease prevalence is primarily correlated with biomass. However, when the degree-biomass relationship is strongly negative, species-level disease prevalence is determined by the difference between a host species’ in-degree and out-degree closeness centrality. At the community level, disease prevalence is generally amplified by increasing host richness, community biomass, and the standard deviation of interaction generality, while it is diluted by higher network connectance. Our framework verifies the amplification effects of host richness during invasion-driven community assembly and offers valuable insights for estimating disease prevalence based on host network topology.

Optimizing kernel width for new risk-sensitive loss: A generalized algorithmic approach

This paper outlines a novel approach to the development of adaptive filtering algorithms, which minimizes susceptibility to the effects of non-Gaussian measurement noise. Using the generalized Gaussian density (GGD) function as the kernel function, we developed a generalized kernel risk-sensitive loss (GKRSL) criterion by which to derive the generalized minimum kernel risk-sensitive loss (GMKRSL) algorithm. We also developed a variable kernel width algorithm for GMKRSL (VKW-GMKRSL) to address the trade-off between convergence rate and steady-state misalignment resulting from a constant kernel width. In simulations, GMKRSL and VKW-GMKRSL outperformed existing works in system identification performance.

Rolling bearing fault diagnosis based on correlated channel attention-optimized convolutional neural networks

Abstract In the field of intelligent fault diagnosis, traditional convolutional neural network (CNN)-based models for rolling bearing fault diagnosis are effective in extracting signal features but fall short in identifying subtle fault features in noisy environments. To address this challenge, this paper introduces a correlated channel attention-optimized deep convolutional neural network (CAOCNN) for fault diagnosis. The main innovations of this study include: firstly, the expansion of the convolutional kernel width through dilated convolution and optimized network parameter settings, which broadens the receptive field for feature extraction and effectively suppresses high-frequency noise; secondly, the relevant channel attention mechanism was constructed., which not only considers the channel weights post-global average pooling but also analyzes the correlations between channel features and the global feature center, dynamically adjusting channel weights to enhance model focus on critical features; additionally, the use of the Nesterov momentum optimization algorithm to optimize network parameters, reducing oscillations and increasing efficiency during training. Experimental results demonstrate that the CAOCNN achieved accuracies of 99.71% and 100% on the Case Western Reserve University and Xi’an Jiaotong University rolling bearing datasets, respectively, improving by 2.91% and 7.6% over traditional CNN models. In noisy conditions, T-SNE visual analysis further confirmed the excellent robustness and feature classification capability of the CAOCNN. These achievements validate the effectiveness of the CAOCNN in diagnosing rolling bearing faults in complex noise environments, contributing valuable advancements to the technology of intelligent fault diagnosis.

Identification and analysis of nine new flo2 allelic mutants in rice

FLO2 is involved in grain development and storage substance synthesis in rice, and therefore can regulate grain size and quality. In this study, we identified 4 new flo2 allelic mutants with nonsense and frameshift mutation in the exon of 6, 10, 11 and 21 and 5 new flo2 allelic mutants with alternative splicing and frameshift mutation at the splicing site of intron 13, 14, 16 and 17. Compared with wild-type rice, the outer endosperm of flo2 mutants was transparent, and the inner endosperm was floury. Different mutation sites and types of FLO2 significantly decreased kernel width, thickness and weight to some extent. The contents of storage protein, starch, amylose and amylopectin showed significant decrease at different levels among 9 flo2 mutants. The expressions of most storage protein synthesis genes and starch synthesis-related genes were significantly down-regulated, and exhibited different ranges of variation among different flo2 mutants. This study could add helpful information for the roles of flo2 alleles in rice quality regulation and provide abundant germplasm resources for rice quality breeding.

Relationships between the number of nuts per cluster and fruit characteristics in the Foşa hazelnut cultivar

This research aimed to determine the effect of the number of nuts per cluster on the fruit characteristics of the Foşa hazelnut cultivar grown in the Arsin district of Trabzon province, Türkiye. Hazelnuts are divided into groups of 1 to 7 according to the number of nuts in cluster. Nut weight, nut length, nut width, nut depth, kernel weight, kernel length, kernel width, kernel depth, shell thickness, kernel ratio, husk length, husk width, and husk opening of the hazelnuts in each group were examined. Principal component analyses (PCA) was used to determine the relationship between the number of nuts in cluster and fruit characteristics. Nut weight, shell thickness, husk length, and husk width values were the highest values in single nut clusters. If the number of nuts in the cluster exceeds four, nut and kernel weights decrease. There was a negative relationship between the nut weight and nut length when the number of nuts in the cluster was above four. A strong relationship were observed in clusters 1, 2, 3, and 4, but a weaker relationship, was seen especially in clusters 5 and 7. A negative relationship was determined between the nut weight and nut length. This study clarified that increasing nut numbers shifts quality characteristics, particularly in a negative trend.

US-LIME: Increasing fidelity in LIME using uncertainty sampling on tabular data

LIME has gained significant attention as an explainable artificial intelligence algorithm that sheds light on how complex machine learning models make decisions within a specific locality. One of the challenges of LIME is its instability and infidelity in acquiring explanations in multiple runs. This study focuses on improving LIME's fidelity by presenting a new sampling strategy. The idea is to generate more focused data samples close to the decision boundary and simultaneously close to the original data point (the sample targeted to be explained). Then, these newly concentrated data are used to train a simple and interpretable linear model as an alternative to the original complex model. The approach leads to high-quality and local sample generation and thus improves the overall fidelity of the model while preserving the constancy of the explanations compared to competing methods. The superiority of the proposed method is shown through comprehensive experiments and comparing the results with LIME, LS-LIME, S-LIME, and BayLIME in terms of fidelity while maintaining stability. This method also performs better than BayLIME, SLIME, and LS-LIME algorithms in terms of execution time. In addition, tests related to the effect of kernel width and data increase on stability and fidelity criteria have been performed. Non-dependence on kernel width in fidelity is also one of the strengths of the proposed method.

Genome-Wide and Transcriptome-Wide Association Analysis Identifies qRS-6D and Its Candidate Genes Regulating Root Development of Wheat Seedlings

Wheat (Triticum aestivum L.) is one of the most important cereal crops worldwide, and its production is challenged by global climate change and a shortage of resources. The root system plays a vital role in uptaking water and nutrients and sensing soil environmental signals, and it has great potential to improve the final yield and stress tolerance of wheat. In order to further explore the genes regulating root development, this study focused on qRS-6D, located on chromosome 6D and spanning from 462,701,391 to 465,068,943, which was significantly associated with the total root length, root volume, root surface, and root fresh weight in our previous GWAS analysis. Firstly, its genetic effects were validated using an F6 segregating population by comparing the root-related traits of homologous lines harboring the alternative haplotypes of this QTL. Then, the number of causal genes of this QTL was narrowed down to four with a transcriptome-wide association study. Additionally, qRS-6D has been demonstrated to have genetic effects on several yield- (kernel length, kernel width, and thousand-kernel weight) and plant structure-related traits (plant height, peduncle length, total tiller number, productive tiller number, flag leaf length, and flag leaf angle). Relatively, the frequency of the favorable haplotype increased with the wheat breeding practice. This study provides a reliable genetic locus to improve root development and structure and evaluate its application potential in wheat breeding improvement.

QTL Mapping of Yield-Related Traits in Tetraploid Wheat Based on Wheat55K SNP Array.

To enhance the understanding of yield-related traits in tetraploid wheat, it is crucial to investigate and identify genes that govern superior yield characteristics. This study utilized the wheat55K single nucleotide polymorphism array to genotype a recombinant inbred line (RIL) population consisting of 120 lines developed through the crossbreeding of two tetraploid wheat varieties, Qin Hei-1 (QH-1) and Durum Wheat (DW). An investigation and analysis were conducted on 11 yield-related traits, including peduncle length (PL), neck length (NL), spike length (SL), flowering date (FD), heading date (HD), thousand-kernel weight (TKW), kernel area ratio (KAR), kernel circumference (KC), kernel length (KL), kernel width (KW), and kernel length-width ratio (KL-WR), over a period of three years in two locations, Yang Ling, Shaanxi, and Lin He, Inner Mongolia. The analysis identified nine stable loci among eight agronomic traits, named QSL.QD-1A.1, QNL.QD-4B.2, QPL.QD-4B.1, QFD.QD-2B, QHD.QD-2B.1, QHD.QD-4B, QKC.QD-4B.2, QKL-WR.QD-4B.6, and QKL.QD-4B.2. Among them, the additive effects of three QTLs, QSL.QD-1A.1, QNL.QD-4B.2, and QFD.QD-2B, were positive, indicating that the enhancing alleles at these loci were derived from the parent line QH-1. These three QTLs showed significant positive effects on the phenotypes of the population materials. Furthermore, potential functional genes were identified within the mapping intervals of QSL.QD-1A.1 and QNL.QD-4B.2, which regulate the development of spike length and neck length, respectively. These results provide potential QTLs and candidate genes, which broaden the genetic basis of agronomic traits related to yield, such as SL, NL, PL, and FD, and benefits for wheat breeding and improvement.

Accelerating the force-coupling method for hydrodynamic interactions in periodic domains

The efficient simulation of fluid-structure interactions at zero Reynolds number requires the use of fast summation techniques in order to rapidly compute the long-ranged hydrodynamic interactions between the structures. One approach for periodic domains involves utilising a compact or exponentially decaying kernel function to spread the force on the structure to a regular grid where the resulting flow and interactions can be computed efficiently using an FFT-based solver. A limitation to this approach is that the grid spacing must be chosen to resolve the kernel and thus, these methods can become inefficient when the separation between the structures is large compared to the kernel width. In this paper, we address this issue for the force-coupling method (FCM) by introducing a modified kernel that can be resolved on a much coarser grid, and subsequently correcting the resulting interactions in a pairwise fashion. The modified kernel is constructed to ensure rapid convergence to the exact hydrodynamic interactions and a positive-splitting of the associated mobility matrix. We provide a detailed computational study of the methodology and establish the optimal choice of the modified kernel width, which we show plays a similar role to the splitting parameter in Ewald summation. Finally, we perform example simulations of rod sedimentation and active filament coordination to demonstrate the performance of fast FCM in application.

MMGAT: a graph attention network framework for ATAC-seq motifs finding.

Motif finding in Assay for Transposase-Accessible Chromatin using sequencing (ATAC-seq) data is essential to reveal the intricacies of transcription factor binding sites (TFBSs) and their pivotal roles in gene regulation. Deep learning technologies including convolutional neural networks (CNNs) and graph neural networks (GNNs), have achieved success in finding ATAC-seq motifs. However, CNN-based methods are limited by the fixed width of the convolutional kernel, which makes it difficult to find multiple transcription factor binding sites with different lengths. GNN-based methods has the limitation of using the edge weight information directly, makes it difficult to aggregate the neighboring nodes' information more efficiently when representing node embedding. To address this challenge, we developed a novel graph attention network framework named MMGAT, which employs an attention mechanism to adjust the attention coefficients among different nodes. And then MMGAT finds multiple ATAC-seq motifs based on the attention coefficients of sequence nodes and k-mer nodes as well as the coexisting probability of k-mers. Our approach achieved better performance on the human ATAC-seq datasets compared to existing tools, as evidenced the highest scores on the precision, recall, F1_score, ACC, AUC, and PRC metrics, as well as finding 389 higher quality motifs. To validate the performance of MMGAT in predicting TFBSs and finding motifs on more datasets, we enlarged the number of the human ATAC-seq datasets to 180 and newly integrated 80 mouse ATAC-seq datasets for multi-species experimental validation. Specifically on the mouse ATAC-seq dataset, MMGAT also achieved the highest scores on six metrics and found 356 higher-quality motifs. To facilitate researchers in utilizing MMGAT, we have also developed a user-friendly web server named MMGAT-S that hosts the MMGAT method and ATAC-seq motif finding results. The advanced methodology MMGAT provides a robust tool for finding ATAC-seq motifs, and the comprehensive server MMGAT-S makes a significant contribution to genomics research. The open-source code of MMGAT can be found at https://github.com/xiaotianr/MMGAT , and MMGAT-S is freely available at https://www.mmgraphws.com/MMGAT-S/ .

Morphological and pomological assessments of seedling-originated walnut (Juglans regia L.) trees to select the promising late-leafing genotypes.

In many parts of the world, including Iran, walnut (Juglans regia L.) production is limited by late-spring frosts. Therefore, the use of late-leafing walnuts in areas with late-spring frost is the most important method to improve yield. In the present study, the phenotypic diversity of 141 seedling genotypes of walnut available in the Senejan area, Arak region, Markazi province, Iran was studied based on morphological traits to obtain superior late-leafing genotypes in the cropping seasons of 2022 and 2023. Based on the results of the analysis of variance, the studied genotypes showed a significant variation in terms of most of the studied morphological and pomological traits. Therefore, it is possible to choose genotypes for different values ​​of a trait. Kernel weight showed positive and significant correlations with leaf length (r = 0.32), leaf width (r = 0.33), petiole length (r = 0.26), terminal leaflet length (r = 0.34), terminal leaflet width (r = 0.21), nut length (r = 0.48), nut width (r = 0.73), nut weight (r = 0.83), kernel length (r = 0.64), and kernel width (r = 0.89). The 46 out of 141 studied genotypes were late-leafing and were analyzed separately. Among late-leafing genotypes, the length of the nut was in the range of 29.33-48.50mm, the width of the nut was in the range of 27.51-39.89mm, and nut weight was in the range of 8.18-16.06g. The thickness of shell was in the range of 1.11-2.60mm. Also, kernel length ranged from 21.97-34.84mm, kernel width ranged from 21.10-31.09mm, and kernel weight ranged from 3.10-7.97g. Based on important and commercial traits in walnut breeding programs, such as nut weight, kernel weight, kernel percentage, kernel color, and ease of kernel removal from nuts, 15 genotypes, including no.92, 91, 31, 38, 33, 18, 93, 3, 58, 108, 16, 70, 15, 82, and 32 were superior and could be used in walnut breeding programs in line with the introduction of new cultivars and the revival of traditional walnut orchards to commercialize them.

Optimized kernel extreme learning machine using Sine Cosine Algorithm for prediction of unconfined compression strength of MICP cemented soil.

Microbially induced calcite precipitation (MICP) is an eco-friendly bio-remediation technology. The unconfined compressive strength (UCS) of MICP cemented soil is an important indicator of repair effectiveness. This study proposes a machine learning technique utilizing the Sine Cosine Algorithm (SCA) to optimize the regularization coefficient C and kernel width γ of the kernel extreme learning machine (KELM) to predict the UCS of MICP cemented soil. To evaluate the performance of the proposed models, a dataset containing 180 groups of the UCS of MICP cemented soil was obtained. The results obtained by SCA-KELM were compared with those obtained by the Random Forest algorithm (RF), Support Vector Machine (SVM), and KELM. The performance of these models was evaluated by the scores of MAE, RMSE, and R2. The results indicate that the SCA-KELM algorithm exhibits optimal prediction performance (Total score: 21). After optimizing KELM with SCA, the total score improved by 110%, suggesting that SCA significantly enhances the KELM performance. After model development, the optimal population size for SCA-KELM was determined to be 50. Based on the mutual information test, an innovative method was developed for categorizing factor sensitivity by employing importance scores as the partitioning criterion. This method categorizes the influencing factors into three tiers: high (importance score: 8.03-11.14%), medium (importance score: 5.93-7.25%), and low (importance score: 3.23-5.18%). These results suggest that the proposed SCA-KELM algorithm can be regarded as a powerful tool for predicting the UCS of MICP cemented soil.

The effective velocity model: An improved approach to velocity sampling in actuator line models

AbstractActuator line modeling of wind turbines requires the definition of a free‐stream velocity in a computational mesh and a regularization kernel to project the computed body forces onto the domain. Both choices strongly influence the results. In this work, a novel velocity sampling method—the so‐called effective velocity model (EVM)—is implemented in the CFD software SOWFA, validated, and compared to pre‐existing approaches. Results show superior method robustness with respect to the regularization kernel width ( ) choice while preserving acceptable accuracy. In particular, the power predicted by the EVM is nearly independent of the value.

Distributed control of partial differential equations using convolutional reinforcement learning

We present a convolutional framework which significantly reduces the complexity and thus, the computational effort for distributed reinforcement learning control of dynamical systems governed by partial differential equations (PDEs). Exploiting translational equivariances, the high-dimensional distributed control problem can be transformed into a multi-agent control problem with many identical, uncoupled agents. Furthermore, using the fact that information is transported with finite velocity in many cases, the dimension of the agents’ environment can be drastically reduced using a convolution operation over the state space of the PDE, by which we effectively tackle the curse of dimensionality otherwise present in deep reinforcement learning. In this setting, the complexity can be flexibly adjusted via the kernel width or by using a stride greater than one (meaning that we do not place an actuator at each sensor location). Moreover, scaling from smaller to larger domains – or the transfer between different domains – becomes a straightforward task requiring little effort. We demonstrate the performance of the proposed framework using several PDE examples with increasing complexity, where stabilization is achieved by training a low-dimensional deep deterministic policy gradient agent using minimal computing resources.

Rolling bearing fault diagnosis based on improved whale-optimization- algorithm–variational-mode-decomposition method

Rolling bearings are a key component of rotating machinery and their health directly affects the safe operation of mechanical equipment. Therefore, fault diagnose for rolling bearings is very important. The fault diagnosis process of rolling bearings includes three stages: signal decomposition, feature extraction, and pattern recognition. Variational mode decomposition (VMD) can suppress end effects, but improper parameter settings will cause information losses or excessive decomposition. In this work, an improved whale optimization algorithm (IWOA) is applied to parameter settings of VMD. Correspondingly, an IWOA-VMD signal decomposition method is proposed. The decomposed signal is combined with a Laplace score method and classifier to remove the redundancy and noise in the feature set and obtain a low-dimensional sensitive feature subset. Then, aiming at the problem of the parameter settings of a least squares support vector machine (LSSVM) affecting the recognition performance and accuracy, a salp swarm algorithm (SSA) is used to globally optimize the penalty parameter and kernel width in the LSSVM to establish an SSA-LSSVM fault recognition model. This model is applied to the fault diagnosis of rolling bearings. In particular, rolling bearing fault samples at Case Western Reserve University are used to verify the method. The results indicate that the proposed method is effective and improves the speed and accuracy of fault diagnosis.

Anomaly detection using large-scale multimode industrial data: An integration method of nonstationary kernel and autoencoder

Kernel methods and neural networks (NNs) are two mainstream nonlinear data modeling methods and have been widely applied to industrial process monitoring. However, they both present imperfect properties, so the relevant applications are limited. On the one hand, kernels are not so reconstructable, scalable, and robust to hyperparameters that they suffer performance degradation for large-scale data modeling and monitoring. On the other hand, the high-dimensional parameter space of NNs that is sorted to parameter initialization presents severe anomaly detection performance inconsistency, which makes the industry cautious about using NNs. Motivated by these facts, we propose to integrate kernels and NNs, forming a new model structure that is scalable, reconstructable, and performance-consistent. Specifically, a novel autoencoder-based nonstationary pattern selection kernel (AE-NPSK) is proposed by (1) selecting from the training set the critical edges and interior data as the centers of the radial basis functions in the hidden layers and (2) adaptively adjusting the kernel width in the training procedure. Also, the new NN has strong performance consistency, which facilitates the search for optimal parameters. Finally, we test the performance of the proposed method on the challenging multimode processes. The results validate the efficacy of the proposed method.

Genome-wide identification and analysis of the GGCT gene family in wheat

Backgroundγ-glutamylcyclotransferase (GGCT), an enzyme to maintain glutathione homeostasis, plays a vital role in the response to plant growth and development as well as the adaptation to various stresses. Although the GGCT gene family analysis has been conducted in Arabidopsis and rice, the family genes have not yet been well identified and analyzed at the genome-wide level in wheat (Triticum aestivum L.).ResultsIn the present study, 20 TaGGCT genes were identified in the wheat genome and widely distributed on chromosomes 2A, 2B, 2D, 3A, 4A, 5A, 5B, 5D, 6A, 6B, 6D, 7A, 7B, and 7D. Phylogenetic and structural analyses showed that these TaGGCT genes could be classified into three subfamilies: ChaC, GGGACT, and GGCT-PS. They exhibited similar motif compositions and distribution patterns in the same subgroup. Gene duplication analysis suggested that the expansion of TaGGCT family genes was facilitated by segmental duplications and tandem repeats in the wheat evolutionary events. Identification of diverse cis-acting response elements in TaGGCT promoters indicated their potential fundamental roles in response to plant development and abiotic stresses. The analysis of transcriptome data combined with RT-qPCR results revealed that the TaGGCTs genes exhibited ubiquitous expression across plant organs, with highly expressed in roots, stems, and developing grains. Most TaGGCT genes were up-regulated after 6 h under 20% PEG6000 and ABA treatments. Association analysis revealed that two haplotypes of TaGGCT20 gene displayed significantly different Thousand-kernel weight (TKW), Kernel length (KL), and Kernel width (KW) in wheat. The geographical and annual distribution of the two haplotypes of TaGGCT20 gene further revealed that the frequency of the favorable haplotype TaGGCT20-Hap-I was positively selected in the historical breeding process of wheat.ConclusionThis study investigated the genome-wide identification, structure, evolution, and expression analysis of TaGGCT genes in wheat. The motifs of TaGGCTs were highly conserved throughout the evolutionary history of wheat. Most TaGGCT genes were highly expressed in roots, stems, and developing grains, and involved in the response to drought stresses. Two haplotypes were developed in the TaGGCT20 gene, where TaGGCT20-Hap-I, as a favorable haplotype, was significantly associated with higher TKW, KL, and KW in wheat, suggesting that the haplotype is used as a function marker for the selection in grain yield in wheat breeding.