Non-destructive time-series assessment of chlorophyll content in flag-leaf(FLC)accurately mimics the senescence rate and the identification of genetic loci associated with senescence provides valuable knowledge to imp...Non-destructive time-series assessment of chlorophyll content in flag-leaf(FLC)accurately mimics the senescence rate and the identification of genetic loci associated with senescence provides valuable knowledge to improve yield stability under stressed environments.In this study,we employed both unmanned aerial vehicles(UAVs)equipped with red–green–blue(RGB)camera and ground-based SPAD-502 instrument to conduct temporal phenotyping of senescence.A total of 262 recombinant inbred lines derived from the cross of Zhongmai 578/Jimai 22 were evaluated for senescence-related traits across three environments,spanning from heading to 35 d post-anthesis.The manual senescence rate(MSR)was quantified using the FLC and the active accumulated temperature,and UAV derived vegetation index were utilized to assess the stay-green rate(USG)facilitating the identification of senescent and stay-green lines.Results indicated that higher senescence rates significantly impacted grain yield,primarily by influencing thousand-kernel weight,and plant height.Quantitative trait loci(QTL)mapping for FLC,USG,and MSR using the 50K SNP array identified 38 stable loci associated with RGB-based vegetation indices and senescence-related traits:among which 19 loci related to senescence traits from UAV and FLC were consistently detected across at least two growth stages,with nine loci likely representing novel QTL.This study highlights the potential of UAV-based high-throughput phenotyping and phenology in identifying critical loci associated with senescence rates in wheat,validating the relationship between senescence rates and yield-related traits in wheat,offering valuable opportunities for gene discovery and significant applications in breeding programs.展开更多
Black point is a black discoloration of the grain embryo that reduces the grain quality and commodity grade.Identifying the underlying genetic loci can facilitate the improvement of black point resistance in wheat.Her...Black point is a black discoloration of the grain embryo that reduces the grain quality and commodity grade.Identifying the underlying genetic loci can facilitate the improvement of black point resistance in wheat.Here,262 recombinant inbred lines(RILs)from the cross of Zhongmai 578/Jimai 22 were evaluated for their black point reactions in fve environments.A high-density genetic linkage map of the RIL population was constructed with the wheat 50K single nucleotide polymorphism(SNP)array.Six stable QTLs for black point resistance were detected,QBp.caas-2A,QBp.caas-2B1,QBp.caas-2B2,QBp.caas-2D,QBp.caas-3A,and QBp.caas-5B,which explained 2.1-28.8%of the phenotypic variances.The resistance alleles of QBp.caas-2B1 and QBp.caas-2B2 were contributed by Zhongmai 578 while the others were from Jimai 22.QBp.caas-2B2,QBp.caas-2D and QBp.caas-3A overlapped with previously reported loci,whereas QBp.caas-2A,QBp.caas-2B1 and QBp.caas-5B are likely to be new.Five kompetitive allele-specifc PCR(KASP)markers,Kasp_2A_BP,Kasp_2B1_BP,Kasp_2B2_BP,Kasp_3A_BP,and Kasp_5B_BP,were validated in a natural population of 165 cultivars.The fndings of this study provide useful QTLs and molecular markers for the improvement of black point resistance in wheat through marker-assisted breeding.展开更多
Wheat(Triticum aestivum)is one of the most essential human energy and protein sources.However,wheat production is threatened by devastating fungal diseases such as stripe rust,caused by Puccinia striiformis Westend.f....Wheat(Triticum aestivum)is one of the most essential human energy and protein sources.However,wheat production is threatened by devastating fungal diseases such as stripe rust,caused by Puccinia striiformis Westend.f.sp.tritici(Pst).Here,we reveal that the alternations in chloroplast lipid profiles and the accumulation of jasmonate(JA)in the necrosis region activate JA signaling and trigger the host defense.The collapse of chloroplasts in the necrosis region results in accumulations of polyunsaturated membrane lipids and the lipid-derived phytohormone JA in transgenic lines of Yr36 that encodes Wheat Kinase START 1(WKS1),a high-temperature-dependent adult plant resistance protein.WKS1.1,a protein encoded by a full-length splicing variant of WKS1,phosphorylates and enhances the activity of keto-acyl thiolase(KAT-2B),a critical enzyme catalyzing theβ-oxidation reaction in JA biosynthesis.The premature stop mutant,kat-2b,accumulates less JA and shows defects in the host defense against Pst.Conversely,overexpression of KAT-2B results in a higher level of JA and limits the growth of Pst.Moreover,JA inhibits the growth and reduces pustule densities of Pst.This study illustrates the WKS1.1-KAT-2B-JA pathway for enhancing wheat defense against fungal pathogens to attenuate yield loss.展开更多
Spike number(SN) per unit area is one of the major determinants of grain yield in wheat. Development of high-throughput techniques to count SN from large populations enables rapid and cost-effective selection and faci...Spike number(SN) per unit area is one of the major determinants of grain yield in wheat. Development of high-throughput techniques to count SN from large populations enables rapid and cost-effective selection and facilitates genetic studies. In the present study, we used a deep-learning algorithm, i.e., Faster Region-based Convolutional Neural Networks(Faster R-CNN) on Red-Green-Blue(RGB) images to explore the possibility of image-based detection of SN and its application to identify the loci underlying SN. A doubled haploid population of 101 lines derived from the Yangmai 16/Zhongmai 895 cross was grown at two sites for SN phenotyping and genotyped using the high-density wheat 660 K SNP array.Analysis of manual spike number(MSN) in the field, image-based spike number(ISN), and verification of spike number(VSN) by Faster R-CNN revealed significant variation(P < 0.001) among genotypes, with high heritability ranged from 0.71 to 0.96. The coefficients of determination(R^(2)) between ISN and VSN was 0.83, which was higher than that between ISN and MSN(R^(2)= 0.51), and between VSN and MSN(R^(2)= 0.50). Results showed that VSN data can effectively predict wheat spikes with an average accuracy of 86.7% when validated using MSN data. Three QTL Qsnyz.caas-4 DS, Qsnyz.caas-7 DS, and QSnyz.caas-7 DL were identified based on MSN, ISN and VSN data, while QSnyz.caas-7 DS was detected in all the three data sets. These results indicate that using Faster R-CNN model for image-based identification of SN per unit area is a precise and rapid phenotyping method, which can be used for genetic studies of SN in wheat.展开更多
Cultivated Triticeae members,including wheat,barley,and rye,are at the center of attention of plant biologists due to their significant contribution to global food security.The completion of the first reference genome...Cultivated Triticeae members,including wheat,barley,and rye,are at the center of attention of plant biologists due to their significant contribution to global food security.The completion of the first reference genome sequence of bread wheat(IWGSC,2018)was a major leap forward for the wheat scientific community.展开更多
Deep learning-based genomic prediction(DL-based GP)has shown promising performance compared to traditional GP methods in plant breeding,particularly in handling large,complex multi-omics data sets.However,the effectiv...Deep learning-based genomic prediction(DL-based GP)has shown promising performance compared to traditional GP methods in plant breeding,particularly in handling large,complex multi-omics data sets.However,the effective development and widespread adoption of DL-based GP still face substantial challenges,including the need for large,high-quality data sets,inconsistencies in performance benchmarking,and the integration of environmental factors.Here,we summarize the key obstacles impeding the development of DL-based GP models and propose future developing directions,such as modular approaches,data augmentation,and advanced attention mechanisms.展开更多
Achieving high yield and good quality in crops is essential for human food security and health.However,there is usually disharmony between yield and quality.Seed storage protein(SSP)and starch,the predominant componen...Achieving high yield and good quality in crops is essential for human food security and health.However,there is usually disharmony between yield and quality.Seed storage protein(SSP)and starch,the predominant components in cereal grains,determine yield and quality,and their coupled synthesis causes a yield–quality trade-off.Therefore,dissection of the underlying regulatory mechanism facilitates simultaneous improvement of yield and quality.Here,we summarize current findings about the synergistic molecular machinery underpinning SSP and starch synthesis in the leading staple cereal crops,including maize,rice and wheat.We further evaluate the functional conservation and differentiation of key regulators and specify feasible research approaches to identify additional regulators and expand insights.We also present major strategies to leverage resultant information for simultaneous improvement of yield and quality by molecular breeding.Finally,future perspectives on major challenges are proposed.展开更多
There is a rapidly rising trend in the development and application of molecular marker assays for gene map- ping and discovery in field crops and trees. Thus far, more than 50 SNP arrays and 15 different types of geno...There is a rapidly rising trend in the development and application of molecular marker assays for gene map- ping and discovery in field crops and trees. Thus far, more than 50 SNP arrays and 15 different types of genotyping-by-sequencing （GBS） platforms have been developed in over 25 crop species and perennial trees. However, much less effort has been made on developing ultra-high-throughput and cost-effective genotyping platforms for applied breeding programs. In this review, we discuss the scientific bottlenecks in existing SNP arrays and GBS technologies and the strategies to develop targeted platforms for crop mo- lecular breeding. We propose that future practical breeding platforms should adopt automated genotyping technologies, either array or sequencing based, target functional polymorphisms underpinning economic traits, and provide desirable prediction accuracy for quantitative traits, with universal applications under wide genetic backgrounds in crops. The development of such platforms faces serious challenges at both the technological level due to cost ineffectiveness, and the knowledge level due to large genotype- phenotype gaps in crop plants. It is expected that such genotyping platforms will be achieved in the next ten years in major crops in consideration of （a） rapid development in gene discovery of important traits, （b） deepened understanding of quantitative traits through new analytical models and population designs, （c） integration of multi-layer -omics data leading to identification of genes and pathways responsible for important breeding traits, and （d） improvement in cost effectiveness of large-scale genotyping. Crop breeding chips and genotyping platforms will provide unprecedented opportunities to accelerate the development of cultivars with desired yield potential, quality, and enhanced adaptation to mitigate the effects of climate change.展开更多
Genomic prediction is an effective way to accelerate the rate of agronomic trait improvement in plants.Traditional methods typically use linear regression models with clear assumptions;such methods are unable to captu...Genomic prediction is an effective way to accelerate the rate of agronomic trait improvement in plants.Traditional methods typically use linear regression models with clear assumptions;such methods are unable to capture the complex relationships between genotypes and phenotypes.Non-linear models(e.g.,deep neural networks)have been proposed as a superior alternative to linear models because they can capture complex non-additive effects.Here we introduce a deep learning(DL)method,deep neural network genomic prediction(DNNGP),for integration of multi-omics data in plants.We trained DNNGP on four datasets and compared its performance with methods built with five classic models:genomic best linear unbiased prediction(GBLUP);two methods based on a machine learning(ML)framework,light gradient boosting machine(LightGBM)and support vector regression(SVR);and two methods based on a DL framework,deep learning genomic selection(DeepGS)and deep learning genome-wide association study(DLGWAS).DNNGP is novel in five ways.First,it can be applied to a variety of omics data to predict phenotypes.Second,the multilayered hierarchical structure of DNNGP dynamically learns features from raw data,avoiding overfitting and improving the convergence rate using a batch normalization layer and early stopping and rectified linear activation(rectified linear unit)functions.Third,when small datasets were used,DNNGP produced results that are competitive with results from the other five methods,showing greater prediction accuracy than the other methods when large-scale breeding data were used.Fourth,the computation time required by DNNGP was comparable with that of commonly used methods,up to 10 times faster than DeepGS.Fifth,hyperparameters can easily be batch tuned on a local machine.Compared with GBLUP,LightGBM,SVR,DeepGS and DLGWAS,DNNGP is superior to these existing widely used genomic selection(GS)methods.Moreover,DNNGP can generate robust assessments from diverse datasets,including omics data,and quickly incorporate complex and large datasets into usable models,making it a promising and practical approach for straightforward integration into existing GS platforms.展开更多
China and CIMMYT have collaborated on wheat improvement for over 40 years and significant progress has been achieved in five aspects in China.A standardized protocol for testing Chinese noodle quality has been establi...China and CIMMYT have collaborated on wheat improvement for over 40 years and significant progress has been achieved in five aspects in China.A standardized protocol for testing Chinese noodle quality has been established with three selection criteria, i.e.,gluten quality, starch viscosity and flour color are identified as being responsible for noodle quality.Genomic approaches have been used to develop and validate genespecific markers, leading to the establishment of a KASP platform, and seven cultivars have been released through application of molecular marker technology.Methodology for breeding adult-plant resistance to yellow rust, leaf rust and powdery mildew, based on the pleiotropic effect of minor genes has been established, resulting in release of six cultivars.More than 330 cultivars derived from CIMMYT germplasm have been released and are now grown over 9% of the Chinese wheat production area.Additionally, physiological approaches have been used to characterize yield potential and develop high-efficiency phenotyping platforms.CIMMYT has also provided valuable training for Chinese scientists.Development of climate-resilient cultivars with application of new technology will be the priority for future collaboration.展开更多
Over the past 70 years,the world has witnessed extraordinary growth in crop productivity,enabled by a suite of technological advances,including higher yielding crop varieties,improved farm management,synthetic agroche...Over the past 70 years,the world has witnessed extraordinary growth in crop productivity,enabled by a suite of technological advances,including higher yielding crop varieties,improved farm management,synthetic agrochemicals,and agricultural mechanization.While this"Green Revolution"intensified crop production,and is credited with reducing famine and malnutrition,its benefits were accompanied by several undesirable collateral effects(Pingali,2012).These include a narrowing of agricultural biodiversity,stemming from increased monoculture and greater reliance on a smaller number of crops and crop varieties for the majority of our calories.This reduction in diversity has created vulnerabilities to pest and disease epidemics,climate variation,and ultimately to human health(Harlan,1972).展开更多
基金funded by the National Key Research and Development Program of China(2022ZD0115703)the National Natural Science Foundation of China(32372196)+1 种基金the Beijing Joint Research Program for Germplasm Innovation and New Variety Breeding(G20220628002)National Natural Science Foundation of China(32250410307)。
文摘Non-destructive time-series assessment of chlorophyll content in flag-leaf(FLC)accurately mimics the senescence rate and the identification of genetic loci associated with senescence provides valuable knowledge to improve yield stability under stressed environments.In this study,we employed both unmanned aerial vehicles(UAVs)equipped with red–green–blue(RGB)camera and ground-based SPAD-502 instrument to conduct temporal phenotyping of senescence.A total of 262 recombinant inbred lines derived from the cross of Zhongmai 578/Jimai 22 were evaluated for senescence-related traits across three environments,spanning from heading to 35 d post-anthesis.The manual senescence rate(MSR)was quantified using the FLC and the active accumulated temperature,and UAV derived vegetation index were utilized to assess the stay-green rate(USG)facilitating the identification of senescent and stay-green lines.Results indicated that higher senescence rates significantly impacted grain yield,primarily by influencing thousand-kernel weight,and plant height.Quantitative trait loci(QTL)mapping for FLC,USG,and MSR using the 50K SNP array identified 38 stable loci associated with RGB-based vegetation indices and senescence-related traits:among which 19 loci related to senescence traits from UAV and FLC were consistently detected across at least two growth stages,with nine loci likely representing novel QTL.This study highlights the potential of UAV-based high-throughput phenotyping and phenology in identifying critical loci associated with senescence rates in wheat,validating the relationship between senescence rates and yield-related traits in wheat,offering valuable opportunities for gene discovery and significant applications in breeding programs.
基金funded by the National Natural Science Foundation of China(32272186)the Beijing Natural Science Foundation,China(6242031)+5 种基金the Basal Research Fund of the Chinese Academy of Agricultural Sciences(CAAS)(S2022QH04)the National Key R&D Program of China(2022YFD1201500)the Young Elite Scientists Sponsorship Program by China Association for Science and Technology(YESS,2020QNRC001)the Modern Cold and Drought Characteristic Agricultural Seed Industry Research Project-2025,Gansu Province,China(ZYGG-2025-8)the Nanfan Special Project,CAAS(YBXM2303)the Science and Technology Innovation Program of CAAS。
文摘Black point is a black discoloration of the grain embryo that reduces the grain quality and commodity grade.Identifying the underlying genetic loci can facilitate the improvement of black point resistance in wheat.Here,262 recombinant inbred lines(RILs)from the cross of Zhongmai 578/Jimai 22 were evaluated for their black point reactions in fve environments.A high-density genetic linkage map of the RIL population was constructed with the wheat 50K single nucleotide polymorphism(SNP)array.Six stable QTLs for black point resistance were detected,QBp.caas-2A,QBp.caas-2B1,QBp.caas-2B2,QBp.caas-2D,QBp.caas-3A,and QBp.caas-5B,which explained 2.1-28.8%of the phenotypic variances.The resistance alleles of QBp.caas-2B1 and QBp.caas-2B2 were contributed by Zhongmai 578 while the others were from Jimai 22.QBp.caas-2B2,QBp.caas-2D and QBp.caas-3A overlapped with previously reported loci,whereas QBp.caas-2A,QBp.caas-2B1 and QBp.caas-5B are likely to be new.Five kompetitive allele-specifc PCR(KASP)markers,Kasp_2A_BP,Kasp_2B1_BP,Kasp_2B2_BP,Kasp_3A_BP,and Kasp_5B_BP,were validated in a natural population of 165 cultivars.The fndings of this study provide useful QTLs and molecular markers for the improvement of black point resistance in wheat through marker-assisted breeding.
基金supported by the National Natural Science Foundation of China(32372557,31972350)the China Postdoctoral Science Foundation(2021M700850)an open project of the State Key Laboratory of Crop Stress Adaptation and Improvement at Henan University,and the Central Government guided Local Science and Technology Development Funds(2023ZY1016).
文摘Wheat(Triticum aestivum)is one of the most essential human energy and protein sources.However,wheat production is threatened by devastating fungal diseases such as stripe rust,caused by Puccinia striiformis Westend.f.sp.tritici(Pst).Here,we reveal that the alternations in chloroplast lipid profiles and the accumulation of jasmonate(JA)in the necrosis region activate JA signaling and trigger the host defense.The collapse of chloroplasts in the necrosis region results in accumulations of polyunsaturated membrane lipids and the lipid-derived phytohormone JA in transgenic lines of Yr36 that encodes Wheat Kinase START 1(WKS1),a high-temperature-dependent adult plant resistance protein.WKS1.1,a protein encoded by a full-length splicing variant of WKS1,phosphorylates and enhances the activity of keto-acyl thiolase(KAT-2B),a critical enzyme catalyzing theβ-oxidation reaction in JA biosynthesis.The premature stop mutant,kat-2b,accumulates less JA and shows defects in the host defense against Pst.Conversely,overexpression of KAT-2B results in a higher level of JA and limits the growth of Pst.Moreover,JA inhibits the growth and reduces pustule densities of Pst.This study illustrates the WKS1.1-KAT-2B-JA pathway for enhancing wheat defense against fungal pathogens to attenuate yield loss.
基金funded by the National Natural Science Foundation of China (31671691, 3171101265, and 31961143007)the National Key Research and Development Program of China(2016YFD0101804)the Fundamental Research Funds for the Institute Planning in Chinese Academy of Agricultural Sciences(S2018QY02)。
文摘Spike number(SN) per unit area is one of the major determinants of grain yield in wheat. Development of high-throughput techniques to count SN from large populations enables rapid and cost-effective selection and facilitates genetic studies. In the present study, we used a deep-learning algorithm, i.e., Faster Region-based Convolutional Neural Networks(Faster R-CNN) on Red-Green-Blue(RGB) images to explore the possibility of image-based detection of SN and its application to identify the loci underlying SN. A doubled haploid population of 101 lines derived from the Yangmai 16/Zhongmai 895 cross was grown at two sites for SN phenotyping and genotyped using the high-density wheat 660 K SNP array.Analysis of manual spike number(MSN) in the field, image-based spike number(ISN), and verification of spike number(VSN) by Faster R-CNN revealed significant variation(P < 0.001) among genotypes, with high heritability ranged from 0.71 to 0.96. The coefficients of determination(R^(2)) between ISN and VSN was 0.83, which was higher than that between ISN and MSN(R^(2)= 0.51), and between VSN and MSN(R^(2)= 0.50). Results showed that VSN data can effectively predict wheat spikes with an average accuracy of 86.7% when validated using MSN data. Three QTL Qsnyz.caas-4 DS, Qsnyz.caas-7 DS, and QSnyz.caas-7 DL were identified based on MSN, ISN and VSN data, while QSnyz.caas-7 DS was detected in all the three data sets. These results indicate that using Faster R-CNN model for image-based identification of SN per unit area is a precise and rapid phenotyping method, which can be used for genetic studies of SN in wheat.
基金the National Natural Science Foundation of China(32361143512)the State Key Laboratory of Crop Gene Resource and Breedingthe National Key R&D Program of China(no.2024YFF1000600).
文摘Cultivated Triticeae members,including wheat,barley,and rye,are at the center of attention of plant biologists due to their significant contribution to global food security.The completion of the first reference genome sequence of bread wheat(IWGSC,2018)was a major leap forward for the wheat scientific community.
基金supported by the National Natural Science Foundation of China(32361143514)Hainan Provincial Natural Science Foundation of China(624MS119)Innovation Program of Chinese Academy of Agricultural Sciences(CAAS-CSIAF-202303).
文摘Deep learning-based genomic prediction(DL-based GP)has shown promising performance compared to traditional GP methods in plant breeding,particularly in handling large,complex multi-omics data sets.However,the effective development and widespread adoption of DL-based GP still face substantial challenges,including the need for large,high-quality data sets,inconsistencies in performance benchmarking,and the integration of environmental factors.Here,we summarize the key obstacles impeding the development of DL-based GP models and propose future developing directions,such as modular approaches,data augmentation,and advanced attention mechanisms.
基金supported by Natural Science Foundation of China(32272182)National Key Research and Development Program of China(2022YFF1002904,2022YFD1201500)+1 种基金STI 2030-Major Projects(2023ZD0406903)the Science and Technology Innovation Program of Chinese Academy of Agricultural Sciences(CAAS)。
文摘Achieving high yield and good quality in crops is essential for human food security and health.However,there is usually disharmony between yield and quality.Seed storage protein(SSP)and starch,the predominant components in cereal grains,determine yield and quality,and their coupled synthesis causes a yield–quality trade-off.Therefore,dissection of the underlying regulatory mechanism facilitates simultaneous improvement of yield and quality.Here,we summarize current findings about the synergistic molecular machinery underpinning SSP and starch synthesis in the leading staple cereal crops,including maize,rice and wheat.We further evaluate the functional conservation and differentiation of key regulators and specify feasible research approaches to identify additional regulators and expand insights.We also present major strategies to leverage resultant information for simultaneous improvement of yield and quality by molecular breeding.Finally,future perspectives on major challenges are proposed.
基金This study was supported by the National Key Research and Development Program of China （2016YFD0101802 and 2016YFE0108600） and National Natural Science Foundation of China （31550110212）.
文摘There is a rapidly rising trend in the development and application of molecular marker assays for gene map- ping and discovery in field crops and trees. Thus far, more than 50 SNP arrays and 15 different types of genotyping-by-sequencing （GBS） platforms have been developed in over 25 crop species and perennial trees. However, much less effort has been made on developing ultra-high-throughput and cost-effective genotyping platforms for applied breeding programs. In this review, we discuss the scientific bottlenecks in existing SNP arrays and GBS technologies and the strategies to develop targeted platforms for crop mo- lecular breeding. We propose that future practical breeding platforms should adopt automated genotyping technologies, either array or sequencing based, target functional polymorphisms underpinning economic traits, and provide desirable prediction accuracy for quantitative traits, with universal applications under wide genetic backgrounds in crops. The development of such platforms faces serious challenges at both the technological level due to cost ineffectiveness, and the knowledge level due to large genotype- phenotype gaps in crop plants. It is expected that such genotyping platforms will be achieved in the next ten years in major crops in consideration of （a） rapid development in gene discovery of important traits, （b） deepened understanding of quantitative traits through new analytical models and population designs, （c） integration of multi-layer -omics data leading to identification of genes and pathways responsible for important breeding traits, and （d） improvement in cost effectiveness of large-scale genotyping. Crop breeding chips and genotyping platforms will provide unprecedented opportunities to accelerate the development of cultivars with desired yield potential, quality, and enhanced adaptation to mitigate the effects of climate change.
基金National Key R&D Program of China(2021YFD1201200)National Science Foundation of China(32022064)+1 种基金Project of Hainan Yazhou Bay Seed Lab(B21HJ0223)Innovation Program of the Chinese Academy of Agricultural Sciences.
文摘Genomic prediction is an effective way to accelerate the rate of agronomic trait improvement in plants.Traditional methods typically use linear regression models with clear assumptions;such methods are unable to capture the complex relationships between genotypes and phenotypes.Non-linear models(e.g.,deep neural networks)have been proposed as a superior alternative to linear models because they can capture complex non-additive effects.Here we introduce a deep learning(DL)method,deep neural network genomic prediction(DNNGP),for integration of multi-omics data in plants.We trained DNNGP on four datasets and compared its performance with methods built with five classic models:genomic best linear unbiased prediction(GBLUP);two methods based on a machine learning(ML)framework,light gradient boosting machine(LightGBM)and support vector regression(SVR);and two methods based on a DL framework,deep learning genomic selection(DeepGS)and deep learning genome-wide association study(DLGWAS).DNNGP is novel in five ways.First,it can be applied to a variety of omics data to predict phenotypes.Second,the multilayered hierarchical structure of DNNGP dynamically learns features from raw data,avoiding overfitting and improving the convergence rate using a batch normalization layer and early stopping and rectified linear activation(rectified linear unit)functions.Third,when small datasets were used,DNNGP produced results that are competitive with results from the other five methods,showing greater prediction accuracy than the other methods when large-scale breeding data were used.Fourth,the computation time required by DNNGP was comparable with that of commonly used methods,up to 10 times faster than DeepGS.Fifth,hyperparameters can easily be batch tuned on a local machine.Compared with GBLUP,LightGBM,SVR,DeepGS and DLGWAS,DNNGP is superior to these existing widely used genomic selection(GS)methods.Moreover,DNNGP can generate robust assessments from diverse datasets,including omics data,and quickly incorporate complex and large datasets into usable models,making it a promising and practical approach for straightforward integration into existing GS platforms.
基金funded by the National Natural Science Foundation of China (31461143021, 31761143006)
文摘China and CIMMYT have collaborated on wheat improvement for over 40 years and significant progress has been achieved in five aspects in China.A standardized protocol for testing Chinese noodle quality has been established with three selection criteria, i.e.,gluten quality, starch viscosity and flour color are identified as being responsible for noodle quality.Genomic approaches have been used to develop and validate genespecific markers, leading to the establishment of a KASP platform, and seven cultivars have been released through application of molecular marker technology.Methodology for breeding adult-plant resistance to yellow rust, leaf rust and powdery mildew, based on the pleiotropic effect of minor genes has been established, resulting in release of six cultivars.More than 330 cultivars derived from CIMMYT germplasm have been released and are now grown over 9% of the Chinese wheat production area.Additionally, physiological approaches have been used to characterize yield potential and develop high-efficiency phenotyping platforms.CIMMYT has also provided valuable training for Chinese scientists.Development of climate-resilient cultivars with application of new technology will be the priority for future collaboration.
文摘Over the past 70 years,the world has witnessed extraordinary growth in crop productivity,enabled by a suite of technological advances,including higher yielding crop varieties,improved farm management,synthetic agrochemicals,and agricultural mechanization.While this"Green Revolution"intensified crop production,and is credited with reducing famine and malnutrition,its benefits were accompanied by several undesirable collateral effects(Pingali,2012).These include a narrowing of agricultural biodiversity,stemming from increased monoculture and greater reliance on a smaller number of crops and crop varieties for the majority of our calories.This reduction in diversity has created vulnerabilities to pest and disease epidemics,climate variation,and ultimately to human health(Harlan,1972).
