Machine learning and deep learning are extensively employed in genomic selection (GS) to expedite the identification of superior genotypes and accelerate breeding cycles. However, a significant challenge with current data-driven deep learning models in GS lies in their low robustness and poor interpretability. To address these challenges, we developed Cropformer, a deep learning framework for predicting crop phenotypes and exploring downstream tasks. This framework combines convolutional neural networks with multiple self-attention mechanisms to improve accuracy. The ability of Cropformer to predict complex phenotypic traits was extensively evaluated on more than 20 traits across five major crops: maize, rice, wheat, foxtail millet, and tomato. Evaluation results show that Cropformer outperforms other GS methods in both precision and robustness, achieving up to a 7.5% improvement in prediction accuracy compared to the runner-up model. Additionally, Cropformer enhances the analysis and mining of genes associated with traits. We identified numerous single nucleotide polymorphisms (SNPs) with potential effects on maize phenotypic traits and revealed key genetic variations underlying these differences. Cropformer represents a significant advancement in predictive performance and gene identification, providing a powerful general tool for improving genomic design in crop breeding. Cropformer is freely accessible at https://cgris.net/ cropformer.

Single-cell RNA sequencing (scRNA-seq) technology enables a deep understanding of cellular differentiation during plant development and reveals heterogeneity among the cells of a given tissue. However, the computational characterization of such cellular heterogeneity is complicated by the high dimensionality, sparsity, and biological noise inherent to the raw data. Here, we introduce PhytoCluster, an unsupervised deep learning algorithm, to cluster scRNA-seq data by extracting latent features. We benchmarked PhytoCluster against four simulated datasets and five real scRNA-seq datasets with varying protocols and data quality levels. A comprehensive evaluation indicated that PhytoCluster outperforms other methods in clustering accuracy, noise removal, and signal retention. Additionally, we evaluated the performance of the latent features extracted by PhytoCluster across four machine learning models. The computational results highlight the ability of PhytoCluster to extract meaningful information from plant scRNA-seq data, with machine learning models achieving accuracy comparable to that of raw features. We believe that PhytoCluster will be a valuable tool for disentangling complex cellular heterogeneity based on scRNA-seq data.

Societal Impact Statement Rice wild relatives (RWR) provide valuable genetic resources for modern rice breeding, yet knowledge gaps constrain their conservation and further utilization. To address these gaps, the potential distributions of 22 RWR taxa were modeled, and their conservation statuses were assessed. Most taxa were identified as medium priority for further conservation. Further ex‐situ collecting hotspots are in Southeast and South Asia, Northern Australia, West Africa, and tropical Americas, while habitat protection lies in Southeast and South Asia, Northern Australia, and West Africa. Climate change may shift equatorial habitats to higher latitudes and mountain habitats to higher altitudes. Summary Rice wild relatives (RWR) provide valuable genetic resources for modern rice breeding. However, knowledge gaps on their geographic distributions and conservation status constrain their conservation and further utilization. To fill these critical gaps, we modeled the potential distributions of 22 RWR taxa under current climate scenarios, assessed their conservation status, both ex situ (in genebanks or botanical gardens) and in situ (in protected areas), and examined changes in taxa richness of RWR in predicted areas under future climate scenarios. The RWR were primarily distributed in tropical Asia to tropical Australia, tropical Africa, and South and Central America. We identified 4 out of 22 taxa as high priority (HP) for further conservation action and 18 taxa as medium priority (MP). Hotspots requiring further collecting for ex situ conservation are concentrated in Southeast and South Asia, Northern Australia, West Africa, and tropical Americas. Meanwhile, habitat protection should be enhanced in Southeast and South Asia, Northern Australia, and West Africa. Under future climate change, suitable habitats near the equator are expected to shift toward higher latitudes and some in lower‐latitude basins may become unsuitable due to excessive heat, resulting in decreased taxa richness in these areas. Additionally, suitable habitats in high mountain areas may shift to higher altitudes, potentially augmenting taxa richness in the highlands. Our findings provide vital insights to guide future rescue conservation efforts for RWR.

The high-quality development of fruit industry is an inevitable requirement to better meet people’s ever-growing needs for a better life, promote farmers’ income, improve agricultural efficiency, and realize the transformation from a big fruit producing country to an industrial power. This paper puts forward the strategic thinking and suggestions of seven aspects to promote the high-quality development of fruit industry in our country: Implement the innovation-driven strategy and accelerate the realization of scientific and technological breakthroughs in the fruit industry; optimize the regional layout of fruit trees, optimize the structure of tree species and varieties; promote the integration of one, two and three industries of fruit industry and realize the industrialization of fruit industry development; standardize the construction of breeding system of fruit trees and cultivate high-quality healthy fruit trees; strengthen early warning of natural disasters,diseases and insect pests and other emergencies, and improve our ability to prevent and mitigate and resist risks; accelerate the demonstration, promotion, transfer and transformation of fruit science and technology achievements, increase the scientific and technological contribution rate of fruit industry; establish of national fruit science and technology innovation alliance to provide strong support for the high-quality development of fruit industry; through the above strategic thinking and suggestions, we hope to provide references for the high quality development of Chinese fruit industry.

从12个方面对我国果业高质量发展的制约因素进行了详细探讨和分析:果树种质资源保护水平和利用质量有待提高,果树区域布局与结构有待优化,果树种苗繁育体系和市场建设滞后,果树自主知识产权品种占比低,果园生产管理技术水平和果品质量不高,突发性自然灾害与病虫害对果树威胁加大,果业绿色发展难度大,果树生产管理机械化和智能化水平低,果品现代冷链物流建设滞后,果品采后商品化处理和加工能力低,果业产业化和组织化程度低,果业技术推广体系急需完善和创新,旨在为我国果业高质量发展措施和发展目标的制定奠定基础.

Background Single-cell RNA sequencing (scRNA-seq) measurements of gene expression show great promise for studying the cellular heterogeneity of rice roots. How precisely annotating cell identity is a major unresolved problem in plant scRNA-seq analysis due to the inherent high dimensionality and sparsity. Results To address this challenge, we present NRTPredictor, an ensemble-learning system, to predict rice root cell stage and mine biomarkers through complete model interpretability. The performance of NRTPredictor was evaluated using a test dataset, with 98.01% accuracy and 95.45% recall. With the power of interpretability provided by NRTPredictor, our model recognizes 110 marker genes partially involved in phenylpropanoid biosynthesis. Expression patterns of rice root could be mapped by the above-mentioned candidate genes, showing the superiority of NRTPredictor. Integrated analysis of scRNA and bulk RNA-seq data revealed aberrant expression of Epidermis cell subpopulations in flooding, Pi, and salt stresses. Conclusion Taken together, our results demonstrate that NRTPredictor is a useful tool for automated prediction of rice root cell stage and provides a valuable resource for deciphering the rice root cellular heterogeneity and the molecular mechanisms of flooding, Pi, and salt stresses. Based on the proposed model, a free webserver has been established, which is available at https://www.cgris.net/nrtp .

Rice landraces, including Asian rice (Oryza sativa L.) and African rice (Oryza glaberrima Steud.), provide important genetic resources for rice breeding to address challenges related to food security. Due to climate change and farm destruction, rice landraces require urgent conservation action. Recognition of the geographical distributions of rice landraces will promote further collecting efforts. Here we modelled the potential distributions of eight rice landrace subgroups using 8351 occurrence records combined with environmental predictors with Maximum Entropy (MaxEnt) algorithm. The results showed they were predicted in eight sub-regions, including the Indus, Ganges, Meghna, Mekong, Yangtze, Pearl, Niger, and Senegal river basins. We then further revealed the changes in suitable areas of rice landraces under future climate change. Suitable areas showed an upward trend in most of study areas, while sub-regions of North and Central China and West Coast of West Africa displayed an unsuitable trend indicating rice landraces are more likely to disappear from fields in these areas. The above changes were mainly determined by changing global temperature and precipitation. Those increasingly unsuitable areas should receive high priority in further collections. Overall, these results provide valuable references for further collecting efforts of rice landraces, while shedding light on global biodiversity conservation.

按照"四个面向"要求,围绕"推进农业高水平科技自立自强、支撑乡村全面振兴"核心使命,根据我国果树产业实际,明确了"保障食物安全与果品有效供给、推动果业绿色高质量发展、推进果品优质安全与营养健康、推动乡村振兴和区域果业发展"等果业重大使命,并提出了我国果业"十四五"的重点任务建议清单.

科技成果有效转移转化,推动科学研究不断创新,支撑产业的高质量发展.国家不断建立健全激励政策,鼓励成果转化模式的不断创新.果树产业是实施乡村振兴战略的支柱产业之一,因其产业自身特点,在科技成果转化中存在难点.以中国农业科学院果树研究所创立的"五五"成果转化模式为研究对象,分析了其创立背景、原则及成效,为农业科技成果转化提供新的思路,不断推动科技创新、成果转化及果业发展质量的变革,助力果树产业高质量可持续发展.

The contributions of crop germplasm resources to food security depend on their conservation and accessibility for use. The automated warehouse has begun to be applied to the ex situ preservation of crop germplasm resources in the crop genebank. Identifying the proper storage scheme for potentially hundreds of thousands of seeds is a primary task that faces the crop genebank manager during the design of a new automated crop genebank. There are mainly three areas to focus on, hardware and software, seeds storage assignment policy and seeds labelling technology. This paper aims to propose automated crop genebank storage schemes for two kinds of crop genebank (the long-term crop genebank and the middle-term crop genebank), which supports managers in determining the technologies that can be applied in the automated crop genebank. Firstly, the selection of hardware and software should be based on the functional orientation of the long-term crop genebank and the middle-term crop genebank. Secondly, for the seed storage assignment policy, the sequential storage assignment is designed for the long-term genebank while the cache storage assignment is developed for the middle-term genebank. Finally, a QR code labelling technology based on image recognition is designed for both the long-term crop genebank and the middle-term crop genebank.

嗜冷黄杆菌(Flavobacterium psychrophilum)是鲑鳟类细菌性冷水病(bacterial cold water disease,BCWD)的病原菌,该病的发生和流行严重制约了鲑鳟产业的健康发展.本研究分析嗜冷黄杆菌肌肉注射感染虹鳟(Oncorhynchus mykiss)后病原菌的动态组织分布情况,以期为BCWD的防控提供理论支撑.以1.0× 108 CFU/mL浓度的嗜冷黄杆菌CH06株菌液肌肉注射感染实验虹鳟,感染后12h、24 h、96 h观察虹鳟临床症状及组织病变,并利用qPCR Taqman探针法检测各组织病原载量.患病鱼的临床病征表现为体色发黑、游动缓慢或不动、食欲不振,尾柄部注射处肌肉溃烂,鳃苍白,脾脏肿大,伴有腹水.组织病理观察显示,嗜冷黄杆菌感染后实验鱼注射处肌纤维断裂、溶解;脾窦扩张,其内充满红细胞;肾脏组织含铁血黄素增加,出现大量空泡变性,肾小管上皮细胞变性、坏死,肾间质炎性细胞浸润.qPCR检测结果表明,肌肉注射感染12 h后即可在脾脏、肝脏、肾脏、肠道、鳃、注射处肌肉、脑和尾鳍中检出嗜冷黄杆菌,注射处肌肉病原载量最高为(5.85±2.11)×105 copy/μL.感染后24 h,脾脏、脑中病原载量较12h时上升最多,其他组织病原载量与感染12h时水平一致,注射处肌肉病原载量为(6.48±2.07)× 105 copy/μL,显著高于其他组织(P＜0.05).感染96 h后脾脏中的病原载量显著高于其他组织(P＜0.05),达到(1.15+0.58)×107 copy/μL;肝脏、肾脏、脾脏中的病原载量较24 h时上升最多.所有被检组织中病原菌载量随时间延长均呈上升趋势.另外,3个时间点注射处肌肉的病原平均载量最高,其次为脾脏,然后是肾脏和鳃.嗜冷黄杆菌人工感染虹鳟后,注射处肌肉、脾脏是细菌的重要增殖场所.综上,嗜冷黄杆菌可随着血液循环进入虹鳟各组织,并表现出对注射处肌肉、脾脏、肾脏和鳃较强的组织嗜性,而对肝脏、尾鳍、肠道和脑组织的嗜性相对较弱:感染时间和病变程度与组织中的病原载量呈正相关.

界定了果业高质量发展的内涵,提出了"一新"驱动、"两优"先行、"三产"融合、"四品"提升、"五减"支撑、"六化"同步的果业高质量发展路径.

为确定患病虹鳟的病原,本实验从患病鱼溃烂肌肉中分离到2株细菌,分别命名为CH06和CH07,经回归感染证实分离菌株为导致此次虹鳟患病的病原菌,并进一步对其形态特征、理化特性、分子特征、血清型及耐药性进行分析.结果显示,CH06和CH07株在TYES琼脂平板上呈煎蛋状外观,产黄色素,氧化酶和过氧化氢酶呈阳性,能水解明胶和酪蛋白,不能水解淀粉,不能利用果糖、半乳糖和七叶苷等.16SrRNA比对结果显示,CH06和CH07株与嗜冷黄杆菌模式株NBRC 15942的同源性分别为99.35％和99.42％.综合菌株理化和分子特性确定CH06和CH07株为嗜冷黄杆菌.利用多重PCR方法鉴定CH06和CH07株的血清型均为1型(Fd型);多位点序列分型(MLST)分析表明,CH06和CH07株的基因型分别为ST-12和ST-78型,且均属于CC-ST10克隆型.人工感染结果显示,CH06和CH07株对虹鳟幼鱼具有较高致病性,其半致死浓度(LD50)分别为7.1×105和1.1×105 CFU/mL,攻毒剂量与临床病症出现时间呈反比,从人工感染实验鱼的肌肉、脾脏等组织中可重新分离到嗜冷黄杆菌.组织病理变化显示,病鱼肝细胞肿胀,空泡变性,部分肝细胞溶解坏死,细胞核溶解消失;脾脏充血、出血,淋巴细胞减少,红细胞和含铁血黄素增多;肌纤维间隙增宽、断裂、弯曲不齐,部分肌细胞肌浆溶解呈蜂窝状.CH06和CH07株对10种抗菌药物的耐药谱略有不同,均对氨苄西林和甲氧苄啶-磺胺甲唑敏感;CH06株对恩诺沙星、氟苯尼考等耐药,而CH07株对恩诺沙星和氟苯尼考中度敏感.本研究首次报道了我国虹鳟源嗜冷黄杆菌的分离鉴定及生物学特性,以期为虹鳟细菌性冷水病的防控提供科学依据.

在分析果业发展不同阶段的基础上,提出了果业5.0,阐明了果业不同阶段的特点,提出推进果业5.0,实现果业高质量发展的建议.

农作物种质资源工作发展至今取得了巨大的成就,但也存在着农作物种质资源家底不清、共享利用效率不高及种质权属不清等问题,开展农作物种质资源登记工作能够有效解决以上问题.本研究从资源登记信息的流向角度出发,梳理了登记的整体工作流程,并结合区块链技术的优势,提出了农作物种质资源登记区块链网络模型.首先根据农作物种质资源登记的业务特点,确定登记主体和登记流程,从数据的填报、审核与共享3个阶段阐明了资源登记的整体流程.在此基础上设计了农作物种质资源登记区块链网络模型、工作流模型和数据模型,并根据农作物种质资源实际工作需求改善了DPOS共识机制,给出了激励机制的构建方式.农作物种质资源登记区块链模型通过去中心化实现登记数据的分布式存储,通过共识机制增加数据的可靠性,通过区块链的特殊数据结构确保链中数据难篡改、易追溯,提高数据安全性,为今后资源登记系统的设计提供了新思路.

为了摸清贵州农业生物资源调查所收集资源的地理分布特征,利用空间统计分析、多样性分析等方法对所收集资源的外在地理环境和内在空间表现进行研究和分析.首先,通过空间叠加和统计确定调查资源所处的海拔、温度、降水等基本地理环境特点;其次,利用优化热点分析、地理分布中心、空间多样性分析等方法分析调查资源在空间上的聚散分布、不同调查专题地理分布的方位和差异性、资源的分布密度、丰富程度等特征.结果 显示,调查资源所处地区的海拔、气象因素的区间跨度较大,但大部分资源处在相对集中的地理环境区间内分布,68.7％的资源分布在海拔600～1400 m之间,65.1％的资源分布在年均温15～17 ℃之间,61.7％的资源分布在年降水1250～1350mm之间;本次收集的资源在省域范围呈现随机分布模式,没有明显的聚集区和离散区域,所选的21个调查县从地理空间的角度上能够基本反映贵州农业生物资源的空间分布特点;不同专题的作物交叉分布,除食用菌集中分布在西南方向外,粮食、经济、蔬菜、果树、药用植物5个专题的资源在省内的分布相对均匀;资源提供者的民族多样,总数多达14个,其中苗族提供资源份数最多,占总数的37％,其次为汉族和布依族,分别为16％和15％,黎族、回族提供者提供的资源比例则低于总数的1％.通过本研究,基本掌握了贵州农业生物资源的分布现状和特征,为贵州省生物资源有效保护和高效利用相关政策的制定提供参考.

本课题组于2012年从中国东部的虹鳟鱼养殖场分离得到了严重威胁我国鲑鳟鱼养殖的传染性造血器官坏死病毒(IHNV)IHNV-Sn1203病毒株.根据GenBank中收录的IHNV病毒株的序列设计引物,将IHNV-Sn1203基因组序列分成5个片段进行RT-PCR克隆,最终得到完整的IHNV-Sn1203基因组序列.利用Lasergene和MEGA 5.0软件分析了IHNV-Sn1203株的全基因组序列、基因组编码蛋白、基因组末端和未翻译序列、以及病毒基因同源性和系统发育.结果 发现,IHNV-Sn1203基因组全长11131nts,共编码6个病毒蛋白,大小分别为核蛋白(N) 1176nts、磷蛋白(P)693nts、基质蛋白(M)588nts、表面糖蛋白(G)1527nts、聚合酶蛋白(L)5961nts和非结构蛋白(NV)336nts.IHNV-Sn1203株各基因间均具有保守的基因终止序列(Gene end,GE)、基因间序列(Intergenic regions,IG)、基因起始序列(Gene star,GS)及Kozak序列,以确保病毒基因的转录和翻译效率.系统进化分析发现,IHNV分为E、U、M、L和J共5个基因型,其中IHNV-Sn1203株与中国其他IHNV株具有显著的亲缘关系,属于J基因型中的J Nagano亚型.由G基因的进化分析发现,J基因型的病毒与U基因型亲缘关系最近,推测IHNV病毒最初可能通过进口鱼卵的方式由U基因型引入,随时间的推移逐渐进化为J基因型.

嗜冷黄杆菌Flavobacterium psychrophilum是细菌性冷水病(Bacterial cold water disease,BCWD)或虹鳟鱼苗综合征(Rainbow trout fry syndrome,RTFS)的病原,主要感染鲑科鱼类幼鱼并可导致较高的死亡率,该菌地理分布广泛,亦可感染非鲑科鱼类并造成疾病.BCWD又称低温病,通常在3～15℃时发生流行,其典型的临床症状包括尾柄部腐蚀、 皮肤溃疡、 鳃苍白、 脾肿大、 腹水和螺旋式游泳行为等.目前,对嗜冷黄杆菌及BCWD的研究主要集中在遗传多样性、 耐药性、 致病性和免疫防控等方面,而关于嗜冷黄杆菌的致病机制仍不够清晰.本文综述了嗜冷黄杆菌的分类地位、 生物学性状、 致病机制,以及BCWD的流行病学和防控技术,以期为BCWD的防控提供科学参考.

Based on understanding the application of big data and the research status of crop germplasm resources, this paper proposes a system architecture that is suitable for crop germplasm resources big data. Among them, the overall architecture of germplasm resources is elaborated through six functional modules, including data source, data integration, data processing, data application, big data operation and maintenance platform, and data management and security. The logical functional architecture specification was formulated and the technical implementation and selection are defined. The technical implementation framework describes the technical implementation of germplasm resources big data, and jointly supports the construction and operation of germplasm resources big data. Finally, a verification system is established to verify the feasibility of the big data system framework for germplasm resources.

农业科研院所是我国农业科技创新体系的重要组成部分.落实新时代发展理念,现代农业科研院所的创新发展是趋势和必然.文章以中国农业科学院郑州果树研究所为例,阐述了2015年以来该所在"顶天立地"的农业科研工作中取得的显著成效,总结了建立产业认知思想体系、现代院所管理体系、科技创新支撑体系、资源投入激励体系、产业服务对接体系等五大体系的工作经验,并提出了下一步创新发展的政策建议.