Plant Methods最新文献

{"title":"Prediction accuracy and repeatability of UAV based biomass estimation in wheat variety trials as affected by variable type, modelling strategy and sampling location.","authors":"Daniel T L Smith, Qiaomin Chen, Sean Reynolds Massey-Reed, Andries B Potgieter, Scott C Chapman","doi":"10.1186/s13007-024-01236-w","DOIUrl":"10.1186/s13007-024-01236-w","url":null,"abstract":"<p><strong>Background: </strong>This study explores the use of Unmanned Aerial Vehicles (UAVs) for estimating wheat biomass, focusing on the impact of phenotyping and analytical protocols in the context of late-stage variety selection programs. It emphasizes the importance of variable selection, model specificity, and sampling location within the experimental plot in predicting biomass, aiming to refine UAV-based estimation techniques for enhanced selection accuracy and throughput in variety testing programs.</p><p><strong>Results: </strong>The research uncovered that integrating geometric and spectral traits led to an increase in prediction accuracy, whilst a recursive feature elimination (RFE) based variable selection workflowled to slight reductions in accuracy with the benefit of increased interpretability. Models, tailored to specific experiments were more accurate than those modelling all experiments together, while models trained for broad-growth stages did not significantly increase accuracy. The comparison between a permanent and a precise region of interest (ROI) within the plot showed negligible differences in biomass prediction accuracy, indicating the robustness of the approach across different sampling locations within the plot. Significant differences in the within-season repeatability (w²) of biomass predictions across different experiments highlighted the need for further investigation into the optimal timing of measurement for prediction.</p><p><strong>Conclusions: </strong>The study highlights the promising potential of UAV technology in biomass prediction for wheat at a small plot scale. It suggests that the accuracy of biomass predictions can be significantly improved through optimizing analytical and modelling protocols (i.e., variable selection, algorithm selection, stage-specific model development). Future work should focus on exploring the applicability of these findings under a wider variety of conditions and from a more diverse set of genotypes.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"129"},"PeriodicalIF":4.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11337646/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142009257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Soybean seed pest damage detection method based on spatial frequency domain imaging combined with RL-SVM.","authors":"Xuanyu Chen, Wei He, Zhihao Ye, Junyi Gai, Wei Lu, Guangnan Xing","doi":"10.1186/s13007-024-01257-5","DOIUrl":"10.1186/s13007-024-01257-5","url":null,"abstract":"<p><p>Soybean seeds are susceptible to damage from the Riptortus pedestris, which is a significant factor affecting the quality of soybean seeds. Currently, manual screening methods for soybean seeds are limited to visual inspection, making it difficult to identify seeds that are phenotypically defect-free but have been punctured by stink bugs on the sub-surface. To facilitate the convenient and efficient identification of healthy soybean seeds, this paper proposes a soybean seed pest detection method based on spatial frequency domain imaging combined with RL-SVM. Firstly, soybean optical data is obtained using single integration sphere technique, and the vigor index of soybean seeds is obtained through germination experiments. Then, based on the above two data items using feature extraction algorithms (the successive projections algorithm and the competitive adaptive reweighted sampling algorithm), the characteristic wavelengths of soybeans are identified. Subsequently, the spatial frequency domain imaging technique is used to obtain the sub-surface images of soybean seeds in a forward manner, and the optical coefficients such as the reduced scattering coefficient <math> <msub><mrow><mi>μ</mi> <mo>'</mo></mrow> <mi>s</mi></msub> </math> and absorption coefficient <math><msub><mi>μ</mi> <mi>a</mi></msub> </math> of soybean seeds are inverted. Finally, RL-MLR, RL-GRNN, and RL-SVM prediction models are established based on the ratio of the area of insect-damaged sub-surface to the entire seed, soybean varieties, and <math><msub><mi>μ</mi> <mi>a</mi></msub> </math> at three wavelengths (502 nm, 813 nm, and 712 nm) for predicting and identifying soybean the stinging and sucking pest damage levels of soybean seeds. The experimental results show that the spatial frequency domain imaging technique yields small errors in the optical coefficients of soybean seeds, with errors of less than 15% for <math><msub><mi>μ</mi> <mi>a</mi></msub> </math> and less than 10% for <math> <msub><mrow><mi>μ</mi> <mo>'</mo></mrow> <mi>s</mi></msub> </math> . After parameter adjustment through reinforcement learning, the Macro-Recall metrics of each model have improved by 10%-15%, and the RL-SVM model achieves a high Macro-Recall value of 0.9635 for classifying the pest damage levels of soybean seeds.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"130"},"PeriodicalIF":4.7,"publicationDate":"2024-08-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11337654/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142009258","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Trans2express - de novo transcriptome assembly pipeline optimized for gene expression analysis.","authors":"Aleksandra M Kasianova, Aleksey A Penin, Mikhail I Schelkunov, Artem S Kasianov, Maria D Logacheva, Anna V Klepikova","doi":"10.1186/s13007-024-01255-7","DOIUrl":"10.1186/s13007-024-01255-7","url":null,"abstract":"<p><strong>Background: </strong>As genomes of many eukaryotic species, especially plants, are large and complex, their de novo sequencing and assembly is still a difficult task despite progress in sequencing technologies. An alternative to genome assembly is the assembly of transcriptome, the set of RNA products of the expressed genes. While a bunch of de novo transcriptome assemblers exists, the challenges of transcriptomes (the existence of isoforms, the uneven expression levels across genes) complicates the generation of high-quality assemblies suitable for downstream analyses.</p><p><strong>Results: </strong>We developed Trans2express - a web-based tool and a pipeline of de novo hybrid transcriptome assembly and postprocessing based on rnaSPAdes with a set of subsequent filtrations. The pipeline was tested on Arabidopsis thaliana cDNA sequencing data obtained using Illumina and Oxford Nanopore Technologies platforms and three non-model plant species. The comparison of structural characteristics of the transcriptome assembly with reference Arabidopsis genome revealed the high quality of assembled transcriptome with 86.1% of Arabidopsis expressed genes assembled as a single contig. We tested the applicability of the transcriptome assembly for gene expression analysis. For both Arabidopsis and non-model species the results showed high congruence of gene expression levels and sets of differentially expressed genes between analyses based on genome and based on the transcriptome assembly.</p><p><strong>Conclusions: </strong>We present Trans2express - a protocol for de novo hybrid transcriptome assembly aimed at recovering of a single transcript per gene. We expect this protocol to promote the characterization of transcriptomes and gene expression analysis in non-model plants and web-based tool to be of use to a wide range of plant biologists.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"128"},"PeriodicalIF":4.7,"publicationDate":"2024-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11330051/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141996305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Melon ripeness detection by an improved object detection algorithm for resource constrained environments.","authors":"Xuebin Jing, Yuanhao Wang, Dongxi Li, Weihua Pan","doi":"10.1186/s13007-024-01259-3","DOIUrl":"10.1186/s13007-024-01259-3","url":null,"abstract":"<p><strong>Background: </strong>Ripeness is a phenotype that significantly impacts the quality of fruits, constituting a crucial factor in the cultivation and harvesting processes. Manual detection methods and experimental analysis, however, are inefficient and costly.</p><p><strong>Results: </strong>In this study, we propose a lightweight and efficient melon ripeness detection method, MRD-YOLO, based on an improved object detection algorithm. The method combines a lightweight backbone network, MobileNetV3, a design paradigm Slim-neck, and a Coordinate Attention mechanism. Additionally, we have created a large-scale melon dataset sourced from a greenhouse based on ripeness. This dataset contains common complexities encountered in the field environment, such as occlusions, overlapping, and varying light intensities. MRD-YOLO achieves a mean Average Precision of 97.4% on this dataset, achieving accurate and reliable melon ripeness detection. Moreover, the method demands only 4.8 G FLOPs and 2.06 M parameters, representing 58.5% and 68.4% of the baseline YOLOv8n model, respectively. It comprehensively outperforms existing methods in terms of balanced accuracy and computational efficiency. Furthermore, it maintains real-time inference capability in GPU environments and demonstrates exceptional inference speed in CPU environments. The lightweight design of MRD-YOLO is anticipated to be deployed in various resource constrained mobile and edge devices, such as picking robots. Particularly noteworthy is its performance when tested on two melon datasets obtained from the Roboflow platform, achieving a mean Average Precision of 85.9%. This underscores its excellent generalization ability on untrained data.</p><p><strong>Conclusions: </strong>This study presents an efficient method for melon ripeness detection, and the dataset utilized in this study, alongside the detection method, will provide a valuable reference for ripeness detection across various types of fruits.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"127"},"PeriodicalIF":4.7,"publicationDate":"2024-08-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11328389/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141996333","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"The effect of testing rate on biomechanical measurements related to stalk lodging.","authors":"Kaitlin Tabaracci, Jacques Vos, Daniel J Robertson","doi":"10.1186/s13007-024-01253-9","DOIUrl":"10.1186/s13007-024-01253-9","url":null,"abstract":"<p><strong>Background: </strong>Stalk lodging (the premature breaking of plant stalks or stems prior to harvest) is a persistent agricultural problem that causes billions of dollars in lost yield every year. Three-point bending tests, and rind puncture tests are common biomechanical measurements utilized to investigate crops susceptibility to lodging. However, the effect of testing rate on these biomechanical measurements is not well understood. In general, biological specimens (including plant stems) are well known to exhibit viscoelastic mechanical properties, thus their mechanical response is dependent upon the rate at which they are deflected. However, there is very little information in the literature regarding the effect of testing rate (aka displacement rate) on flexural stiffness, bending strength and rind puncture measurements of plant stems.</p><p><strong>Results: </strong>Fully mature and senesced maize stems and wheat stems were tested in three-point bending at various rates. Maize stems were also subjected to rind penetration tests at various rates. Testing rate had a small effect on flexural stiffness and bending strength calculations obtained from three-point bending tests. Rind puncture measurements exhibited strong rate dependent effects. As puncture rate increased, puncture force decreased. This was unexpected as viscoelastic materials typically show an increase in resistive force when rate is increased.</p><p><strong>Conclusions: </strong>Testing rate influenced three-point bending test results and rind puncture measurements of fully mature and dry plant stems. In green stems these effects are expected to be even larger. When conducting biomechanical tests of plant stems it is important to utilize consistent span lengths and displacement rates within a study. Ideally samples should be tested at a rate similar to what they would experience in-vivo.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"125"},"PeriodicalIF":4.7,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11323486/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141982987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Segmentation and characterization of macerated fibers and vessels using deep learning.","authors":"Saqib Qamar, Abu Imran Baba, Stéphane Verger, Magnus Andersson","doi":"10.1186/s13007-024-01244-w","DOIUrl":"10.1186/s13007-024-01244-w","url":null,"abstract":"<p><strong>Purpose: </strong>Wood comprises different cell types, such as fibers, tracheids and vessels, defining its properties. Studying cells' shape, size, and arrangement in microscopy images is crucial for understanding wood characteristics. Typically, this involves macerating (soaking) samples in a solution to separate cells, then spreading them on slides for imaging with a microscope that covers a wide area, capturing thousands of cells. However, these cells often cluster and overlap in images, making the segmentation difficult and time-consuming using standard image-processing methods.</p><p><strong>Results: </strong>In this work, we developed an automatic deep learning segmentation approach that utilizes the one-stage YOLOv8 model for fast and accurate segmentation and characterization of macerated fiber and vessel form aspen trees in microscopy images. The model can analyze 32,640 x 25,920 pixels images and demonstrate effective cell detection and segmentation, achieving a <math><msub><mtext>mAP</mtext> <mrow><mn>0.5</mn> <mo>-</mo> <mn>0.95</mn></mrow> </msub> </math> of 78 %. To assess the model's robustness, we examined fibers from a genetically modified tree line known for longer fibers. The outcomes were comparable to previous manual measurements. Additionally, we created a user-friendly web application for image analysis and provided the code for use on Google Colab.</p><p><strong>Conclusion: </strong>By leveraging YOLOv8's advances, this work provides a deep learning solution to enable efficient quantification and analysis of wood cells suitable for practical applications.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"126"},"PeriodicalIF":4.7,"publicationDate":"2024-08-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11325806/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141983017","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An optimised CRISPR Cas9 and Cas12a mutagenesis toolkit for Barley and Wheat.","authors":"Tom Lawrenson, Martha Clarke, Rachel Kirby, Macarena Forner, Burkhard Steuernagel, James K M Brown, Wendy Harwood","doi":"10.1186/s13007-024-01234-y","DOIUrl":"10.1186/s13007-024-01234-y","url":null,"abstract":"<p><strong>Background: </strong>CRISPR Cas9 and Cas12a are the two most frequently used programmable nucleases reported in plant systems. There is now a wide range of component parts for both which likely have varying degrees of effectiveness and potentially applicability to different species. Our aim was to develop and optimise Cas9 and Cas12a based systems for highly efficient genome editing in the monocotyledons barley and wheat and produce a user-friendly toolbox facilitating simplex and multiplex editing in the cereal community.</p><p><strong>Results: </strong>We identified a Zea mays codon optimised Cas9 with 13 introns in conjunction with arrayed guides driven by U6 and U3 promoters as the best performer in barley where 100% of T0 plants were simultaneously edited in all three target genes. When this system was used in wheat > 90% of T0 plants were edited in all three subgenome targets. For Cas12a, an Arabidopsis codon optimised sequence with 8 introns gave the best editing efficiency in barley when combined with a tRNA based multiguide array, resulting in 90% mutant alleles in three simultaneously targeted genes. When we applied this Cas12a system in wheat 86% & 93% of T0 plants were mutated in two genes simultaneously targeted. We show that not all introns contribute equally to enhanced mutagenesis when inserted into a Cas12a coding sequence and that there is rationale for including multiple introns. We also show that the combined effect of two features which boost Cas12a mutagenesis efficiency (D156R mutation and introns) is more than the sum of the features applied separately.</p><p><strong>Conclusion: </strong>Based on the results of our testing, we describe and provide a GoldenGate modular cloning system for Cas9 and Cas12a use in barley and wheat. Proven Cas nuclease and guide expression cassette options found in the toolkit will facilitate highly efficient simplex and multiplex mutagenesis in both species. We incorporate GRF-GIF transformation boosting cassettes in wheat options to maximise workflow efficiency.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"123"},"PeriodicalIF":4.7,"publicationDate":"2024-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11321142/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141976271","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Orchid2024: A cultivar-level dataset and methodology for fine-grained classification of Chinese Cymbidium Orchids.","authors":"Yingshu Peng, Yuxia Zhou, Li Zhang, Hongyan Fu, Guimei Tang, Guolin Huang, Weidong Li","doi":"10.1186/s13007-024-01252-w","DOIUrl":"10.1186/s13007-024-01252-w","url":null,"abstract":"<p><strong>Background: </strong>Chinese Cymbidium orchids, cherished for their deep-rooted cultural significance and significant economic value in China, have spawned a rich tapestry of cultivars. However, these orchid cultivars are facing challenges from insufficient cultivation practices and antiquated techniques, including cultivar misclassification, complex identification, and the proliferation of counterfeit products. Current commercial techniques and academic research primarily emphasize species identification of orchids, rather than delving into that of orchid cultivars within species.</p><p><strong>Results: </strong>To bridge this gap, the authors dedicated over a year to collecting a cultivar image dataset for Chinese Cymbidium orchids named Orchid2024. This dataset contains over 150,000 images spanning 1,275 different categories, involving visits to 20 cities across 12 provincial administrative regions in China to gather pertinent data. Subsequently, we introduced various visual parameter-efficient fine-tuning (PEFT) methods to expedite model development, achieving the highest top-1 accuracy of 86.14% and top-5 accuracy of 95.44%.</p><p><strong>Conclusion: </strong>Experimental results demonstrate the complexity of the dataset while highlighting the considerable promise of PEFT methods within flower image classification. We believe that our work not only provides a practical tool for orchid researchers, growers and market participants, but also provides a unique and valuable resource for further exploring fine-grained image classification tasks. The dataset and code are available at https://github.com/pengyingshu/Orchid2024 .</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"124"},"PeriodicalIF":4.7,"publicationDate":"2024-08-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11320945/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141976272","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Advancing virus-induced gene silencing in sunflower: key factors of VIGS spreading and a novel simple protocol","authors":"Majd Mardini, Mikhail Kazancev, Elina Ivoilova, Victoria Utkina, Anastasia Vlasova, Yakov Demurin, Alexander Soloviev, Ilya Kirov","doi":"10.1186/s13007-024-01241-z","DOIUrl":"https://doi.org/10.1186/s13007-024-01241-z","url":null,"abstract":"Virus-Induced Gene Silencing (VIGS) is a versatile tool in plant science, yet its application to non-model species like sunflower demands extensive optimization due to transformation challenges. In this study, we aimed to elucidate the factors that significantly affect the efficiency of Agrobacterium-VIGS in sunflowers. After testing a number of approaches, we concluded that the seed vacuum technique followed by 6 h of co-cultivation produced the most efficient VIGS results. Genotype-dependency analysis revealed varying infection percentages (62–91%) and silencing symptom spreading in different sunflower genotypes. Additionally, we explored the mobility of tobacco rattle virus (TRV) and phenotypic silencing manifestation (photo-bleaching) across different tissues and regions of VIGS-infected sunflower plants. We showed the presence of TRV is not necessarily limited to tissues with observable silencing events. Finally, time-lapse observation demonstrated a more active spreading of the photo-bleached spots in young tissues compared to mature ones. This study not only offers a robust VIGS protocol for sunflowers but also provides valuable insights into genotype-dependent responses and the dynamic nature of silencing events, shedding light on TRV mobility across different plant tissues.","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"86 1","pages":""},"PeriodicalIF":5.1,"publicationDate":"2024-08-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141933799","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of deep learning for predicting rice traits using structural and single-nucleotide genomic variants.","authors":"Ioanna-Theoni Vourlaki, Sebastián E Ramos-Onsins, Miguel Pérez-Enciso, Raúl Castanera","doi":"10.1186/s13007-024-01250-y","DOIUrl":"10.1186/s13007-024-01250-y","url":null,"abstract":"<p><strong>Background: </strong>Structural genomic variants (SVs) are prevalent in plant genomes and have played an important role in evolution and domestication, as they constitute a significant source of genomic and phenotypic variability. Nevertheless, most methods in quantitative genetics focusing on crop improvement, such as genomic prediction, consider only Single Nucleotide Polymorphisms (SNPs). Deep Learning (DL) is a promising strategy for genomic prediction, but its performance using SVs and SNPs as genetic markers remains unknown.</p><p><strong>Results: </strong>We used rice to investigate whether combining SVs and SNPs can result in better trait prediction over SNPs alone and examine the potential advantage of Deep Learning (DL) networks over Bayesian Linear models. Specifically, the performances of BayesC (considering additive effects) and a Bayesian Reproducible Kernel Hilbert space (RKHS) regression (considering both additive and non-additive effects) were compared to those of two different DL architectures, the Multilayer Perceptron, and the Convolution Neural Network, to explore their prediction ability by using various marker input strategies. We found that exploiting structural and nucleotide variation slightly improved prediction ability on complex traits in 87% of the cases. DL models outperformed Bayesian models in 75% of the studied cases, considering the four traits and the two validation strategies used. Finally, DL systematically improved prediction ability of binary traits against the Bayesian models.</p><p><strong>Conclusions: </strong>Our study reveals that the use of structural genomic variants can improve trait prediction in rice, independently of the methodology used. Also, our results suggest that Deep Learning (DL) networks can perform better than Bayesian models in the prediction of binary traits, and in quantitative traits when the training and target sets are not closely related. This highlights the potential of DL to enhance crop improvement in specific scenarios and the importance to consider SVs in addition to SNPs in genomic selection.</p>","PeriodicalId":20100,"journal":{"name":"Plant Methods","volume":"20 1","pages":"121"},"PeriodicalIF":4.7,"publicationDate":"2024-08-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11316328/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"141913682","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"生物学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}