im冷钱包下载|nvwa
Nature Genetics | 基于人工智能神经网络的基因组解读系统Nvwa并揭示细胞命运决定共性规律 - 知乎
Nature Genetics | 基于人工智能神经网络的基因组解读系统Nvwa并揭示细胞命运决定共性规律 - 知乎首发于Life Science & AI切换模式写文章登录/注册Nature Genetics | 基于人工智能神经网络的基因组解读系统Nvwa并揭示细胞命运决定共性规律DrugAIAI赋能药物研发与医疗本文介绍由浙江大学基础医学院的郭国骥、韩晓平和良渚实验室的王晶晶共同通讯发表在 Nature Genetics 的研究成果:目前研究人员在生成和分析基因组方面做了大量努力,但大多数物种仍缺乏预测基因调控和细胞命运决定的遗传模型。在该研究中,作者利用自主构建的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并探究了八种代表性的后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型保守的调控程序。作者开发了一种基于深度学习的模型Nvwa,用于在单细胞分辨率下预测基因表达和识别调控序列。作者还系统地比较了细胞类型特异性转录因子,以揭示脊椎动物和无脊椎动物细胞类型的保守遗传调控。该工作有助于为研究不同生物系统的调控语法提供宝贵的资源和新的策略。简介单细胞是生命的基本单位。高通量单细胞RNA测序(scRNA-seq)增强了研究人员识别细胞类型的能力。随着scRNA-seq技术的发展,scRNA-seq分析已被用于绘制各种物种的全生物体细胞图谱,包括人类、斑马鱼、果蝇、小鼠、线虫和涡虫。细胞类型是多细胞生命的基本组成部分,由转录因子(TF)等核心调控因子控制。最近,细胞类型被提出作为具有准独立进化变化潜力的“进化单位”。具有共同谱系祖先的细胞类型共享核心调控TF,其可能在物种进化过程中发生分化。单细胞图谱为系统比较不同物种的细胞类型和调节因子提供了前所未有的机会。虽然TF的表达可以用scRNA-seq来测量,但目前尚不清楚基因组是如何在细胞图谱背后编码不同的时空遗传程序的。深度神经网络是建模高维数据中复杂关系的强大方法,有助于学习在特定条件下从基因组序列到基因表达的映射。目前已经开发了几种模型来预测DNA序列中的基因表达或染色质谱,如Xpresso、DeepSEA、Basset、 Enformer和AI-TAC。这些深度学习模型在识别复杂序列模式方面显示出了强大的能力。然而,此类模型尚未应用于多物种的综合图谱,并且细胞图谱水平的深度神经网络有可能识别出跨生物共享的新细胞类型特异性调控。在该研究中作者构建了斑马鱼、果蝇和蚯蚓的全身单细胞图谱。并收集了八种代表性的后生动物图谱,研究了细胞类型和TF的跨物种相似性。然后,作者开发了一种基于深度学习的模型Nvwa,以从单个细胞的DNA序列预测基因表达。最后,作者还解释了细胞类型特异性的序列规则,并表征了跨物种细胞类型的保守调控程序。结果构建斑马鱼、果蝇和蚯蚓的单细胞图谱之前,作者使用Microwell-seq构建了人类和小鼠的全生物体细胞图谱。在这项研究中,作者使用可以消除组织特异性批次效应的scRNA-seq策略构建了斑马鱼、果蝇和蚯蚓的全身细胞图谱(图1a)。其中,斑马鱼图谱收集了635,228个单细胞数据,果蝇图谱涵盖了276,706个单细胞数据,蚯蚓图谱包含了95,020个单细胞数据。通过对图谱数据进行无监督聚类,发现了105种主要的斑马鱼细胞类型(图1b)、87种主要的果蝇细胞类型(图1c)和62种主要的蚯蚓细胞类型,它们具有不同的基因表达程序。作者还根据典型细胞类型特异性标记的标准化表达水平对每个细胞类型进行注释。总共105种斑马鱼细胞被分为11个主要细胞谱系:内皮细胞、上皮细胞、红系细胞、生殖细胞、肝细胞、免疫细胞、肌肉细胞、神经元细胞、分泌细胞、基质细胞和其他细胞。作者还对105种主要细胞类型中的每一种进行了子聚类分析,并在层次结构中识别出1285个细胞类型子聚类(图1d)。果蝇的细胞图谱中,87种细胞类型被分为12个主要的细胞谱系:上皮细胞、神经元细胞、血细胞、卵泡、肠细胞、生殖细胞、雄性副腺、马氏小管(MT)、肌肉细胞、增殖细胞、脂肪体和其他细胞。同时,作者使用MetaNeighbor将构建的单细胞图谱与一个平行的蝇细胞图谱项目进行了比较。在87种果蝇细胞类型中,约93.1%与组织特异性注释一致。最后,对87种主要细胞类型中的每一种进行子聚类分析,在层次结构中共识别出1085个子聚类(图1e)。图1 使用Microwell-seq构建斑马鱼和果蝇细胞图谱对于蚯蚓细胞图谱,62种细胞类型被分为8个主要的细胞谱系:消化腺细胞、上皮细胞、神经元细胞、体腔细胞、肌肉细胞、红细胞、生殖细胞和其他细胞。最后,子聚类分析在蚯蚓层次结构中共识别出462个子聚类,这些子聚类在功能上是有意义的。斑马鱼、果蝇和蚯蚓的图谱资源可在http://bis.zju.edu.cn/nvwa/获得。此外,作者还在结构细胞(非免疫细胞)中观察到显著的免疫基因活性,包括斑马鱼上皮细胞、果蝇MT细胞和蚯蚓体腔细胞。并且作者证明了哺乳动物的结构细胞,包括上皮细胞、内皮细胞和基质细胞,可能具有免疫特征,从而有助于提高对造血系统外的免疫学理解。细胞图谱中的跨物种比较利用流式细胞术和群体分析在生物体水平上研究和建模基因调控模式一直是一个挑战。通过统一的单细胞信使RNA测序(mRNA-seq)平台,作者构建的细胞图谱数据资源为研究跨物种细胞分类的遗传调控提供了前所未有的机会。因此,作者旨在分析细胞类型特异性遗传调控网络,并通过数据集成和机器学习来评估跨物种遗传调控的保守性。为了获得高质量的细胞,作者设置了一个更高的截止值,以生成一个数据集,其中斑马鱼和果蝇平均每个细胞大约有1000个基因,蚯蚓平均每个细胞大约有400个基因。作者总共从八种物种中获得了480种细胞类型,涵盖了主要的细胞谱系,包括上皮细胞、免疫细胞、神经元细胞、基质细胞、肌肉细胞、分泌细胞、红系细胞、生殖细胞、内皮细胞和增殖细胞谱系。然后使用伪细胞算法制作伪体细胞计数矩阵或基于马尔可夫亲和力的细胞图插补 (MAGIC),以插补缺失的基因表达。图2 八个物种的跨物种分析为了检验细胞类型的跨物种相似性,对八个转录组数据集进行了成对SAMap分析。作者比较了MAGIC、单细胞和伪细胞三种数据集的跨物种比较结果,结果发现脊椎动物的细胞类型是保守的。基于MAGIC数据集,85.9%同源细胞类型对可以基于单细胞和伪细胞数据集重新识别。为了降低结果的假阳性率,作者设置了严格的阈值来构建跨物种图谱。经分析可知脊椎动物的细胞类型是保守的,特别是免疫细胞、基质细胞、神经元细胞、上皮细胞、内皮细胞和生殖细胞(图2a)。在果蝇的跨物种图谱中,作者还发现几乎来自同一细胞谱系的所有细胞类型都显示出很强的联系(图2b)。为了进一步验证跨物种图谱的结果,作者对肌肉和神经元中同源细胞类型之间的富集基因对进行了功能富集分析(图2c),作者发现富集的基因对具有一致功能,这与之前的研究结果是一致的。为了评估脊椎动物和无脊椎动物在调控水平上的调控保守性和细胞类型差异,作者计算了每个物种的TF特异性得分(图3a-h)。总的来说,作者在八个物种中共鉴定出2342个细胞谱系特异性TF。基于八个物种间同源基因的转换,可以观察到同源TF中更多保守特征。同源TF分别覆盖了人类、小鼠和斑马鱼所有细胞类型特异性TF的91.42%(70个中的64个)、98.75%(80个中的79个)和75%(104个中的78个)。总之,作者的研究为保守遗传调控基因的跨物种筛选提供了保守遗传调控的详细信息。图3 利用scRNA-seq数据比较物种内部和物种间的调节TFNvwa根据DNA序列预测基因表达TF作为调控网络中的重要功能节点,可以识别特定的DNA序列来控制染色质状态和转录。然而,确定DNA序列元件与细胞类型特异性基因调控相关的分子机制仍然具有挑战性。为了更好地理解基因组中编码的调控元件,作者开发了一种基于深度学习的模型Nvwa。训练Nvwa模型以从单热调控序列的输入中预测单个细胞各基因的表达。对预测的细胞图谱进行评估,以验证模型的性能。此后,将学习的序列规则以调控元件的形式进行解释,如序列基序及其预测影响。Nvwa配备了单细胞分辨率,可以进一步识别特定细胞类型与深度学习衍生序列基序之间的关联(图4a)。综上所述,Nvwa可以仅在单细胞水平上利用基因调控序列预测基因表达并识别特定于细胞类型的候选调控因子。作者首先独立训练了八个物种的Nvwa模型,并评估了Nvwa能否正确预测单细胞基因表达。Nvwa表达预测的准确度是通过检测数据中受试者操作特征曲线(AUROC)下的平均面积和精确召回曲线(AUPR)下的面积来评估的。Nvwa稳健地预测了八个物种的基因表达,其总体AUROC为0.78,AUPR为0.59。通过比较不同细胞类型的性能,表达预测正确性最高的总是生殖系的细胞。Nvwa在预测单细胞基因表达方面进行了优化,在人类和果蝇数据集中优于Basset、DeepSEA、Beluga和Basenji等标准架构。此外,通过集成相关物种的序列进行多基因组训练,可以进一步提高Nvwa模型的准确性。Nvwa模型预测再现了细胞之间的关系,包括细胞类型的相似性和多样性,预测结果与在同一细胞类型中观察到的表达更为相似。细胞类型特异性进一步通过t分布随机邻居嵌入(t-SNE)和预测表达位点在保留基因上的调整互信息(AMI)评分得到证实。总的来说,这些评价证实了Nvwa可以从DNA序列中正确预测单细胞水平的基因表达。Nvwa可以进一步扩展到扫描全基因组转录活性信号,尽管只训练了基因调控序列(平均约占基因组的13%)。Nvwa模型沿着整个染色体扫描序列,通过识别调控DNA序列来预测信号。通过检查Nvwa全基因组预测,作者观察到它们与实验测量的功能基因组数据相关。此外,通过可视化基因组浏览器轨迹,可观察到Nvwa预测与多种细胞类型和物种中实验定义的信号之间的一致性(图4b)。总的来说,实验分析从外部验证了Nvwa预测性能的鲁棒性。在其应用中,Nvwa模型可以作为在硅片中进行功能基因组研究的辅助工具。图4 深度学习模型框架的应用和解释Nvwa确定特定细胞类型的调控程序为了理解为什么Nvwa可以正确预测单细胞基因表达,作者检查了学习模型的过滤器,其代表了对相应细胞类型重要的特定序列基序。正如预期的那样,在TSS周围的窗口内系统地移动输入序列表明,近端启动子区域(±1 kbp)的信息量最大。然后,作者基于特征映射和TF-MoDISco方法从每个第一层卷积滤波器中提取深度学习的基序,并发现这两种方法给出了一致的结果。作者还计算了序列基序的细胞类型特异性,并使用影响评分进行量化。结果表明,与不同TF相关的过滤器也参与了细胞类型的识别和细胞活性(图4c)。这些结果启发作者进一步分析模型过滤器及其与细胞类型特异性基序和TF的关系。Nvwa衍生的序列基序可以分配到已知的TF结合位点(TFBS)。作者还观察到,带注释的滤波器与已知的TFBS高度相似(图4d)。在交叉验证分析中,大多数注释滤波器具有较高的再现性和信息含量,这表明Nvwa解释的鲁棒性。一些影响分数较高的未注释过滤器可能捕获了较短的序列模式。除了生物学注释,作者还检查了Nvwa序列基序的细胞类型特异性。对于小鼠和果蝇,50%-80%的细胞类型特异性Nvwa基序通过相应的单细胞ATAC-seq数据被重新识别。作者还发现,细胞类型特异性过滤器与相应TF的已知作用一致(图4e,f)。作者还在果蝇中鉴定了过滤调节子,这证实了由相同过滤器调控的靶基因具有相似的细胞谱系特异性表达模式(图5a,b)。总之,这些结果表明,Nvwa可以利用与特定细胞类型相关的TF的深度学习衍生基序,使得能够直接从序列中筛选细胞类型特异性调控因子。图5 Nvwa 确定特定细胞类型的调控程序Nvwa基序的跨物种比较为了进一步分析物种间的遗传网络,作者比较了基于深度学习的基序在物种间的保守性和差异性。作者在八个物种特异性模型中共识别出663个细胞类型特异性过滤器。约94.9%的细胞类型特异性过滤器至少与来自其他物种特异性模型的一个过滤器同源(图6)。并且同源过滤器倾向于保持物种间相似的细胞类型特异性。深度学习基序的跨物种比较显示出揭示特定细胞类型下保守调控因子富集的潜力。图6 细胞谱系特异性过滤器的保守水平分布总结在这项研究中,作者利用自主构建的高通量单细胞测序平台Microwell-seq构建了斑马鱼、果蝇和蚯蚓的全生物细胞图谱。在没有组织特异性批次效应的情况下测量了整个细胞的平衡状态。这些细胞图谱为研究物种,特别是节肢动物和环节动物的细胞分类提供了前所未有的机会。在这项研究中,作者总共分析了八种具有代表性的后生动物物种,以生成一个详细目录,来说明动物进化过程中细胞平衡状态的保守性和多样性。为了区分趋同进化和协同进化,作者筛选了具有细胞类型特异性的细胞谱系特异性TF。该研究为在单细胞分辨率下深入理解比较基因组学提供了一个框架。未来的研究可能会收集更多的后生动物物种,以追踪重要细胞类型的出现和研究细胞进化。同时,作者开发了一个基于深度学习的框架Nvwa,仅从DNA序列预测细胞图谱水平的基因表达。Nvwa具有与特定细胞状态相关的预测调控功能,这使作者能够直接从序列中筛选细胞类型特异性的调控因子。此外,Nvwa仅使用基因组序列就可以模拟多细胞生物的复杂表达模式。Nvwa从未使用任何表观基因组数据进行训练,但其全基因组活性预测与使用功能基因组学确定的候选调控元件相关。这些结果有两个含义。首先,利用基因组共享的基本规则,深度神经网络可以模拟多细胞基因表达图谱。第二,谱系特异性转录组在很大程度上由调控DNA序列决定。虽然Nvwa为研究进化过程中细胞类型特异性调控程序提供了一个新的视角,但Nvwa模型的解释和应用仍需谨慎。首先,超参数和模型体系结构,特别是第一层卷积滤波器控制了序列模式解释的简并性和灵敏度之间的权衡,应该根据用户的特定目的进行调整。例如,可以增加滤波器数量以提高序列基序检测的灵敏度。其次,使用Nvwa进行全基因组预测可以帮助研究人员进行功能基因组研究,并填充高度重复的基因组区域。但由于Nvwa尚处于概念验证阶段,其预测结果与具体实验数据并不完全一致;因此,Nvwa应该在实践中作为辅助工具使用。第三,本研究聚焦于TF调控因子,并将深度学习衍生的序列模式解释为TF基序。然而,仍然有新的序列模式不能分配到已知的数据库。第四,通过功能实验验证调控元件非常重要。最后,基因调控机制复杂,模型的体系结构、预测性能和调控逻辑解释仍有待完善。总之,作者生成了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并开发了一种基于深度学习的模型Nvwa,来预测基因表达并识别单细胞水平的调控序列,作者还揭示了进化过程中保守调控程序的作用。该研究将为破解多物种调控图谱提供宝贵的资源。参考资料Li, J., Wang, J., Zhang, P. et al. Deep learning of cross-species single-cell landscapes identifies conserved regulatory programs underlying cell types. Nat Genet (2022).https://doi.org/10.1038/s41588-022-01197-7数据https://figshare.com/s/ecc05b1051fb5678fd3ehttp://bis.zju.edu.cn/nvwa/代码https://github.com/JiaqiLiZju/Nvwa/发布于 2022-10-29 16:40深度学习(Deep Learning)人工智能生物信息学和计算生物学赞同 6添加评论分享喜欢收藏申请转载文章被以下专栏收录Life Science & AI生命科学和生物信息与人工智能交叉领域
女娲计划
女娲计划
首页
目标范围
漏洞提交
流程图
问答(Q&A)
EN
"女娲计划"是关于0day漏洞及利用技术研究奖励计划。主要针对主流的PC及移动操作系统,流行服务端或客户端软件应用,网络设备,虚拟系统逃逸等0day安全漏洞及相关利用技术研究项目提供丰厚的奖金,最高单个漏洞可达¥20,000,000。
目标范围
包括但不限于以下范围,目标范围会定期更新
最新目标列表new
目标
版本
类型
要求
日期
有效期(d)
最高单价(人民币)
长期目标
目标
版本
类型
要求
日期
有效期(d)
最高单价(人民币)
3CX
-
RCE
零交互
2020-09-04
永久
150,000
ABB
-
RCE
零交互
2020-09-04
永久
150,000
ABB Ability
-
RCE
零交互
2020-09-04
永久
150,000
Acme
-
RCE
零交互
2020-09-04
永久
150,000
ActiveMQ
-
RCE
零交互
2021-01-05
永久
50,000
ADC
-
RCE
零交互
2021-01-05
永久
20,000
Adobe
-
RCE
一次交互
2020-09-04
永久
800,000
Adobe-PDF
-
RCE
一次交互
2019-11-08
永久
800,000
advantech
-
RCE
零交互
2020-09-04
永久
150,000
Airos
-
RCE
零交互
2020-09-04
永久
350,000
Android
-
FCWP
零交互
2019-11-08
永久
20,000,000
Anydesk
-
RCE
零交互
2020-09-04
永久
150,000
anymacro安宁邮箱
-
RCE/AUL
零交互
2021-01-05
永久
500,000
Apache
-
RCE
零交互
2020-09-04
永久
500,000
Apache Shiro
-
RCE
零交互
2021-01-05
永久
100,000
Apache Spark
-
RCE
零交互
2021-01-05
永久
20,000
Apache Struts2
-
RCE
零交互
2021-01-05
永久
500,000
Apereo CAS
-
RCE/AUL
零交互
2021-01-05
永久
100,000
ArcSight
-
RCE
零交互
2020-08-07
永久
350,000
ASUS
-
RCE
零交互
2020-09-04
永久
200,000
Atlassian
-
RCE
零交互
2020-09-04
永久
400,000
Atlassian Jira
-
RCE
零交互
2021-01-05
永久
300,000
AudioCodes
-
RCE
零交互
2020-09-04
永久
150,000
Avaya
-
RCE
零交互
2020-09-04
永久
150,000
Barracuda
-
RCE
零交互
2020-09-04
永久
200,000
beyondtrust
-
RCE
零交互
2020-09-04
永久
200,000
BlueCoat
-
RCE
零交互
2020-09-04
永久
200,000
Bluecoat-ProxySG
-
RCE
零交互
2019-11-08
永久
800,000
BroadForward
-
RCE
零交互
2020-09-04
永久
200,000
Cacti
-
RCE
零交互
2020-09-04
永久
150,000
CheckPoint
-
RCE
零交互
2020-09-04
永久
350,000
Chrome
-
RCE+LPE
零交互
2019-11-08
永久
3,000,000
Cisco
-
RCE
零交互
2019-11-08
永久
5,000,000
CISCO firewall
-
RCE
零交互
2021-01-05
永久
500,000
CISCO SSL VPN
-
RCE/AUL
零交互
2021-01-05
永久
500,000
Citrix
-
RCE
零交互
2020-09-04
永久
350,000
ClearScada
-
RCE
零交互
2020-09-04
永久
150,000
Confluence
-
RCE/AUL
零交互
2021-01-05
永久
300,000
Coremail
-
RCE/AUL
零交互
2021-01-05
永久
500,000
CourierMail Server
-
RCE
零交互
2020-09-04
永久
150,000
Cpanel
-
RCE
零交互
2020-08-07
永久
350,000
Cyberoam
-
RCE
零交互
2020-09-04
永久
200,000
Dedecms
-
RCE
零交互
2021-01-05
永久
50,000
Defense Software
-
RCE
零交互
2019-11-08
永久
500,000
Diameteriq
-
RCE
零交互
2020-09-04
永久
150,000
Discuz
-
RCE
零交互
2021-01-05
永久
50,000
dlink
-
RCE
零交互
2021-01-05
永久
50,000
Docker
-
SBX
零交互
2020-09-04
永久
500,000
Dovecot
-
RCE
零交互
2020-09-04
永久
150,000
Drupal
-
RCE
零交互
2020-09-04
永久
150,000
easysite
-
RCE
零交互
2021-01-05
永久
30,000
ECShop
-
RCE
零交互
2021-01-05
永久
20,000
Emerson
-
RCE
零交互
2020-09-04
永久
150,000
EmpireCMS
-
RCE
零交互
2021-01-05
永久
20,000
Ericsson HSS
-
RCE
零交互
2020-09-04
永久
150,000
eScan
-
RCE
零交互
2020-09-04
永久
150,000
Exchange
-
RCE
零交互
2020-08-07
永久
800,000
EXIM
-
RCE
零交互
2020-09-04
永久
350,000
express
-
RCE
零交互
2021-01-05
永久
50,000
F5
-
RCE
零交互
2020-09-04
永久
500,000
F5 BIG-IP
-
RCE
零交互
2021-01-05
永久
500,000
Fastjson
-
RCE
零交互
2021-01-05
永久
500,000
Firefox
-
RCE+LPE
零交互
2019-11-01
永久
800,000
FortiGate
-
RCE
零交互
2020-08-07
永久
350,000
Fortigate-Firewall
-
RCE
零交互
2019-11-08
永久
800,000
FortiNet
-
RCE
零交互
2020-09-04
永久
350,000
Fortinet(飞塔) Firewall
-
RCE
零交互
2021-01-05
永久
50,000
Foxit
-
RCE+LPE
一次交互
2020-09-04
永久
500,000
FreeBSD
-
LPE
零交互
2020-08-07
永久
500,000
FusionAccess
-
RCE
零交互
2021-01-05
永久
50,000
Gitea
-
RCE
零交互
2021-01-05
永久
50,000
Gitlab
-
RCE
零交互
2021-01-05
永久
50,000
Grandstream
-
RCE
零交互
2020-09-04
永久
150,000
H3C
-
RCE
零交互
2020-08-07
永久
500,000
Hadoop
-
RCE
零交互
2021-01-05
永久
50,000
HanSight Enterprise
-
RCE
零交互
2021-01-05
永久
50,000
Harbor
-
RCE
零交互
2021-01-05
永久
20,000
HttpFileServer
-
RCE
零交互
2020-09-04
永久
150,000
IBM
-
RCE
零交互
2020-09-04
永久
350,000
IE
-
RCE
一次交互
2020-09-04
永久
800,000
Ignition
-
RCE
零交互
2020-09-04
永久
150,000
IIS
-
RCE
零交互
2020-09-04
永久
3,000,000
iOS
-
FCWP
零交互
2019-11-08
永久
15,000,000
jackson
-
RCE
零交互
2021-01-05
永久
500,000
Jboss
-
RCE
零交互
2020-09-04
永久
500,000
jeecms
-
RCE
零交互
2021-01-05
永久
20,000
JeeSite
-
RCE
零交互
2021-01-05
永久
10,000
Jenkins
-
RCE
零交互
2021-01-05
永久
50,000
jetty
-
RCE
零交互
2021-01-05
永久
300,000
JFinal
-
RCE
零交互
2021-01-05
永久
20,000
jumpserver
-
RCE/AUL
零交互
2021-01-05
永久
50,000
juniper
-
RCE
零交互
2019-11-08
永久
2,000,000
Kaspersky
-
RCE
零交互
2020-09-04
永久
150,000
kxmail
-
RCE/AUL
零交互
2021-01-05
永久
50,000
Laravel
-
RCE
零交互
2021-01-05
永久
20,000
Liferay
-
RCE
零交互
2020-09-04
永久
150,000
Linksys
-
RCE
零交互
2020-09-04
永久
150,000
Linux
-
LPE
零交互
2019-11-08
永久
500,000
MAC
-
RCE
零交互
2020-09-04
永久
3,000,000
Mailman
-
RCE
零交互
2020-09-04
永久
150,000
McAfee
-
RCE
零交互
2020-09-04
永久
200,000
MetInfo
-
RCE
零交互
2021-01-05
永久
10,000
Microsoft
-
RCE
零交互
2020-09-04
永久
1,000,000
Microsoft SharePoint
-
RCE
零交互
2021-01-05
永久
200,000
Mikrotik
-
RCE
零交互
2020-09-04
永久
150,000
ModSecurity
-
RCE
零交互
2021-01-05
永久
20,000
MOXA
-
RCE
零交互
2020-09-04
永久
150,000
MS-Office
-
RCE
零交互
2019-11-08
永久
1,500,000
NAGIOS
-
RCE
零交互
2020-09-04
永久
200,000
Netflow
-
RCE
零交互
2020-09-04
永久
150,000
NetScreen
-
RCE
零交互
2020-09-04
永久
150,000
Nexus
-
RCE
零交互
2021-01-05
永久
20,000
Onlyoffice
-
RCE
零交互
2020-09-04
永久
150,000
OpenFind
-
RCE
零交互
2020-09-04
永久
150,000
OSPF Routing Protocol
-
RCE
零交互
2020-09-04
永久
150,000
Other Office
-
RCE
零交互
2019-11-08
永久
500,000
Outlook
-
RCE
零交互
2020-09-04
永久
1,500,000
Paloalto
-
RCE
零交互
2020-09-04
永久
150,000
Peplink
-
RCE
零交互
2020-09-04
永久
150,000
PFsense
-
RCE
零交互
2020-09-04
永久
350,000
phabricator
-
RCE
零交互
2020-09-04
永久
150,000
PHP
-
RCE
零交互
2020-09-04
永久
1,500,000
Phpcms
-
RCE
零交互
2021-01-05
永久
50,000
Phpmyadmin
-
RCE
零交互
2021-01-05
永久
100,000
phpStudy
-
RCE
零交互
2021-01-05
永久
20,000
PLESK
-
RCE
零交互
2020-09-04
永久
800,000
Profibus protocol
-
RCE
零交互
2020-09-04
永久
150,000
Pulse Secure
-
RCE
零交互
2020-09-04
永久
350,000
Pulse Secure VPN
-
RCE/AUL
零交互
2021-01-05
永久
500,000
QEMU
-
VME
零交互
2020-08-07
永久
800,000
Qnap
-
RCE
零交互
2020-09-04
永久
150,000
Redmine
-
RCE
零交互
2020-09-04
永久
150,000
resin
-
RCE
零交互
2021-01-05
永久
100,000
Ribbon
-
RCE
零交互
2020-09-04
永久
150,000
richmail(thinkmail)
-
RCE/AUL
零交互
2021-01-05
永久
100,000
RoundCube
-
RCE
零交互
2020-09-04
永久
150,000
SaltStack
-
RCE
零交互
2021-01-05
永久
20,000
Sangoma
-
RCE
零交互
2020-09-04
永久
150,000
Schneider
-
RCE
零交互
2020-08-07
永久
350,000
SE Inno CMS
-
RCE
零交互
2020-09-04
永久
150,000
SendMail
-
RCE
零交互
2020-09-04
永久
1,000,000
SharePoint
-
RCE
零交互
2020-09-04
永久
200,000
Siemens
-
RCE
零交互
2020-09-04
永久
200,000
SIMATIC
-
RCE
零交互
2020-09-04
永久
200,000
SiteServer
-
RCE
零交互
2021-01-05
永久
20,000
SNMP
-
RCE
零交互
2020-09-04
永久
500,000
Solaris
-
LPE
零交互
2019-11-08
永久
500,000
Solarwinds
-
RCE
零交互
2020-09-04
永久
150,000
Sonus
-
RCE
零交互
2020-09-04
永久
150,000
Sophos
-
RCE
零交互
2020-09-04
永久
200,000
Splunk
-
RCE
零交互
2020-09-04
永久
150,000
Spring Boot
-
RCE
零交互
2021-01-05
永久
500,000
Spring Security Oauth
-
RCE
零交互
2021-01-05
永久
50,000
StormShield
-
RCE
零交互
2020-09-04
永久
200,000
Struts2
-
RCE
零交互
2020-09-04
永久
350,000
SWIFTNet
-
RCE
零交互
2020-09-04
永久
150,000
Symantec
-
RCE
零交互
2020-09-04
永久
150,000
Synology
-
RCE
零交互
2020-09-04
永久
150,000
TACACS
-
RCE
零交互
2020-09-04
永久
150,000
TeamViewer
-
RCE
零交互
2021-01-05
永久
50,000
Telegram
-
RCE+LPE
零交互
2019-11-08
永久
5,000,000
Thinkphp
-
RCE
零交互
2020-09-04
永久
500,000
TPlink
-
RCE
零交互
2021-01-05
永久
50,000
Trend Micro
-
RCE
零交互
2020-09-04
永久
200,000
turbomail
-
RCE/AUL
零交互
2021-01-05
永久
20,000
Unify
-
RCE
零交互
2020-09-04
永久
150,000
Virtual Box
-
VME
零交互
2020-09-04
永久
500,000
Vmware
-
VME
零交互
2020-09-04
永久
800,000
VMware ESXi
-
VME
零交互
2019-11-08
永久
1,500,000
VMware vCenter
-
RCE
零交互
2021-01-05
永久
100,000
VMware Workstation
-
VME
零交互
2019-11-08
永久
600,000
Vnc Viewer Server
-
FCWP
零交互
2020-05-29
永久
500,000
VxWorks
-
VME
零交互
2020-09-04
永久
800,000
WatchGuard
-
RCE
零交互
2020-09-04
永久
150,000
WebEOC
-
RCE
零交互
2020-09-04
永久
150,000
Weblogic
-
RCE
零交互
2020-09-04
永久
350,000
Webmin
-
RCE
零交互
2021-01-05
永久
50,000
websphere
-
RCE
零交互
2021-01-05
永久
300,000
-
RCE+LPE
零交互
2019-11-08
永久
10,000,000
Whatsup Gold
-
RCE
零交互
2020-09-04
永久
150,000
Windows
-
RCE
零交互
2019-11-01
永久
10,000,000
winmail
-
RCE/AUL
零交互
2021-01-05
永久
350,000
Winrar
-
RCE
一次交互
2020-09-04
永久
500,000
Wordpress
-
RCE
零交互
2020-09-04
永久
500,000
XAMPP
-
RCE
零交互
2021-01-05
永久
20,000
Yeastar
-
RCE
零交互
2020-09-04
永久
150,000
Zabbix
-
RCE
零交互
2020-08-07
永久
500,000
Zimbra
-
RCE
零交互
2020-09-04
永久
350,000
Zoho
-
RCE
零交互
2020-09-04
永久
150,000
万户ezoffice
-
RCE
零交互
2021-01-05
永久
20,000
亿邮
-
RCE/AUL
零交互
2021-01-05
永久
50,000
向日葵
-
ALL
零交互
2021-01-05
永久
50,000
堡垒机
-
RCE/AUL
零交互
2021-01-05
永久
50,000
大汉cms
-
RCE
零交互
2021-01-05
永久
20,000
宝塔
-
RCE
零交互
2021-01-05
永久
20,000
帕拉迪堡垒机
-
RCE/AUL
零交互
2021-01-05
永久
50,000
常用安防类产品(防火墙、VPN、IDS、IPS、主机安全、终端安全等)
-
RCE
零交互
2021-01-06
永久
50,000
微擎
-
RCE
零交互
2021-01-05
永久
10,000
拓尔思 TRSWAS
-
RCE
零交互
2021-01-05
永久
20,000
日志易
-
RCE
零交互
2021-01-05
永久
50,000
时代亿信邮箱
-
RCE/AUL
零交互
2021-01-05
永久
200,000
泛微
-
RCE/AUL
零交互
2021-01-05
永久
50,000
爱快流控路由
-
RCE
零交互
2021-01-05
永久
50,000
用友
-
RCE/AUL
零交互
2021-01-05
永久
50,000
禅知
-
RCE
零交互
2021-01-05
永久
20,000
禅道/zentao
-
RCE
零交互
2021-01-05
永久
50,000
税友
-
RCE
零交互
2021-01-05
永久
50,000
致远oa
-
RCE/AUL
零交互
2021-01-05
永久
50,000
蓝凌
-
RCE/AUL
零交互
2021-01-05
永久
50,000
通达oa
-
RCE/AUL
零交互
2021-01-05
永久
20,000
金蝶
-
RCE
零交互
2021-01-05
永久
50,000
锐捷
-
RCE
零交互
2021-01-05
永久
50,000
齐治堡垒机
-
RCE/AUL
零交互
2021-01-05
永久
100,000
ALL:RCE + LPE;RCE(Remote Code Execution):远程代码执行;LPE(Local
Privilege Escalation):本地权限提升;SBX(Sandbox Escape Bypass):沙盒逃逸绕过;VME(Virtual Machine Escape):虚拟机逃逸;FCWP(Full
Chain (Zero-Click) with Persistence):完整的利用链;AUL:任意用户登录
提交漏洞
你的邮箱
目标名称+版本+操作系统及版本+体系结构(32位/64位/都可以)(目标名称请使用原表格内的名词)
攻击场景说明(如需要打开文档或者打开网页等)
自测成功率
漏洞利用后是否crash
目标默认配置是否触发(如果不是默认配置,请在下面说明条件)
漏洞触发是否需要身份认证或者凭证权限说明
触发漏洞是否需要交互及说明
漏洞成功利用后的权限说明
其他说明
自评价格(期望获取的奖励额度)
您的公开PGP密钥(建议使用)
验证码
提交漏洞
问答
1.邮箱
请使用您的常用邮箱,在漏洞确认期间,我们会通过邮件和您联系。
2.漏洞接收范围?
我们重点主要关注"女娲计划"给出的目标范围内的相关0day漏洞,目标范围相关的漏洞收取会定期更新,请随时关注目标列表。
您的漏洞在我们给出的目标范围之外,如果漏洞影响大并且严重,也可能成为我们的接收目标,可直接提交,我们评估后联系回复您。
3.我能得到多少漏洞奖金?
在您提交漏洞简介时会一并提交漏洞自评价格,我们会根据您提交的漏洞信息评估您的漏洞价值并和您联系议价。只有在议价得到双方认可之后,流程才会进入到下一步。
在接收到漏洞样本详情之后,如果我们发现与您之前的描述不符,我们会根据您提及的漏洞细节进行二次议价。
4.我需要提交什么样的漏洞简介?
在流程开始阶段,请参考"漏洞提交"表单信息提交,到确定漏洞议价完成后需要提交漏洞详情说明及完整可靠的exploit。
5.在提交漏洞之后,我什么时候能拿到奖金?
在提交完整漏洞说明及exploit后,经过平台确认完整可靠后我们将根据最终议价结果发放奖金。奖金将在3个月内完成分期支付。
最终确定漏洞后1周内支付完成50%奖金,其余50%奖金作为保证金最终在漏洞确认后3个月内完成100%支付。
漏洞提交者应该为漏洞相关信息及程序保密,如果是由于漏洞提交者原因出现泄密情况我们会根据对应具体情况进行扣除或者取消奖励。
6."要求"内的交互相关要求(如"零交互")是什么意思?
交互相关要求是指攻击场景之外的其他任何交互动作,比如打开office打开文档或者浏览器打开链接、移动im等应用打开链接都算是"攻击场景",不属于"交互要求"范围内。
7.在沟通过程中,如果需要发送我认为敏感的信息怎么办?
在沟通过程中,如果你担心你发送的信息涉及到敏感信息,请使用我们的公开PGP密钥加密。
我们的PGP公钥如下:
public-key-service@nvwa.org
8.如果我有问题想要咨询怎么办?
如果有任何问题,请发送邮件到 root#nvwa.org。值得注意的是,这个邮箱不会沟通任何于实际漏洞相关的信息,当你提交表单之后,请于联系你的邮箱进行进一步沟通。
NÜWA:女娲算法,多模态预训练模型,大杀四方!(附源代码下载) - 知乎
NÜWA:女娲算法,多模态预训练模型,大杀四方!(附源代码下载) - 知乎首发于计算机视觉研究院切换模式写文章登录/注册NÜWA:女娲算法,多模态预训练模型,大杀四方!(附源代码下载)计算机视觉研究院西安电子科技大学 电子与通信工程硕士计算机视觉研究院卫星公众号 ID|ComputerVisionGzq学习群|在主页获取加入方式论文地址:https://arxiv.org/abs/2111.12417源代码:https:// http://github.com/microsoft/NUWA计算机视觉研究院专栏作者:Edison_G最近看到一篇论文,名字首先吸引了,内容大概看了后,觉得还是不错的,今天有幸给大家慢慢分享,有兴趣的同学可以阅读论文,深入继续了解!一、前言今天分享的论文,主要提出了一个统一的多模态预训练模型,称为NÜWA,可以为各种视觉合成任务生成新的或操纵现有的视觉数据(即图像和视频)。针对不同场景同时覆盖语言、图像和视频,设计了3D Transformer编码器-解码器框架,不仅可以将视频作为3D数据处理,还可以分别将文本和图像作为1D和2D数据进行适配。还提出了3D Nearby Attention(3DNA)机制来考虑视觉数据的性质并降低计算复杂度。在8个下游任务上评估NÜWA。与几个强大的基线相比,NÜWA在文本到图像生成、文本到视频生成、视频预测等方面取得了最先进的结果。此外,它还显示了令人惊讶的良好的文本零样本能力——引导图像和视频处理任务。8个任务的案例二、背景如今,网络变得比以往任何时候都更加视觉化,图像和视频已成为新的信息载体,并已被用于许多实际应用中。在此背景下,视觉合成正成为越来越受欢迎的研究课题,其目的是构建可以为各种视觉场景生成新的或操纵现有视觉数据(即图像和视频)的模型。自回归模型【Auto-regressive models】在视觉合成任务中发挥着重要作用,因为与GAN相比,它们具有显式的密度建模和稳定的训练优势。早期的视觉自回归模型,如PixelCNN、PixelRNN、Image Transformer、iGPT和Video Transformer,都是以“pixel-by-pixel”的方式进行视觉合成的。然而,由于它们在高维视觉数据上的高计算成本,这些方法只能应用于低分辨率的图像或视频,并且难以扩展。最近,随着VQ-VAE作为离散视觉标记化方法的出现,高效和大规模的预训练可以应用于图像的视觉合成任务(例如DALL-E和CogView) 和视频(例如GODIVA)。尽管取得了巨大的成功,但此类解决方案仍然存在局限性——它们分别处理图像和视频,并专注于生成它们中的任何一个。这限制了模型从图像和视频数据中受益。三、NÜWA的表现Text-To-Image(T2I)一只戴着护目镜,盯着摄像机的狗Sketch-To-Image (S2I)草图转图片任务,就是根据草图的布局,生成对应的图片Image Completion (I2I)图像补全,如果一副图片残缺了,算法可以自动“脑补”出残缺的部分Image Manipulation (TI2I)图片处理,根据文字描述,处理图片例如:有一副草原的图片,然后增加一段描述:一匹马奔跑在草原上,然后就可以生成对应的图片。Video四、新框架NÜWA模型的整体架构包含一个支持多种条件的 adaptive 编码器和一个预训练的解码器,能够同时使图像和视频的信息。对于图像补全、视频预测、图像处理和视频处理任务,将输入的部分图像或视频直接送入解码器即可。而编码解码器都是基于一个3D NEARBY SELF-ATTENTION(3DNA)建立的,该机制可以同时考虑空间和时间轴的上局部特性,定义如下:W 表示可学习的权重,X 和 C 分别代表文本、图像、视频数据的 3D 表示。3DNA考虑了完整的邻近信息,并为每个token动态生成三维邻近注意块。注意力矩阵还显示出3DNA的关注部分(蓝色)比三维块稀疏注意力和三维轴稀疏注意力更平滑。3D DATA REPRESENTATION为了涵盖所有文本、图像和视频或其草图,研究者将它们全部视为标记并定义统一的 3D符号X∈Rh×w×s×d,其中h和w表示空间轴(分别为高度和宽度)中的标记数量,s表示时间轴上的标记数量,d是每个标记的维度。3D NEARBY SELF-ATTENTION基于之前的3D数据表示定义了一个统一的3D Nearby Self-Attention (3DNA) 模块,支持自注意力和交叉注意力。首先给出方程中3DNA的定义:并在如下等式中介绍详细的实现。3D ENCODER-DECODER开始介绍基于3DNA构建的3D编码-解码器。为了在C∈Rh′×w′×s′×din的条件下生成目标Y∈Rh×w×s×dout,Y和C的位置编码通过考虑高度、宽度和时间轴的三个不同的可学习词汇更新。然后,条件C被输入到具有L 3DNA层堆栈的编码器中,以对自注意力交互进行建模,第l层在等式中表示:同样,解码器也是一堆L 3DNA层。解码器计算生成结果的自注意力以及生成结果和条件之间的交叉注意力。第l层表示如下等式。五、实验简单分析其他实验可在论文中获取!© The Ending转载请联系本号获得授权计算机视觉研究院学习群等你加入!计算机视觉研究院主要涉及深度学习领域,主要致力于人脸检测、人脸识别,多目标检测、目标跟踪、图像分割等研究方向。研究院接下来会不断分享最新的论文算法新框架,我们这次改革不同点就是,我们要着重”研究“。之后我们会针对相应领域分享实践过程,让大家真正体会摆脱理论的真实场景,培养爱动手编程爱动脑思考的习惯!关注计算机视觉研究院ID|ComputerVisionGzq学习群|在主页获取加入方式 往期推荐 实用教程详解:模型部署,用DNN模块部署YOLOv5目标检测(附源代码)LCCL网络:相互指导博弈来提升目标检测精度(附源代码)Poly-YOLO:更快,更精确的检测(主要解决Yolov3两大问题,附源代码)ResNet超强变体:京东AI新开源的计算机视觉模块!(附源代码)Double-Head:重新思考检测头,提升精度(附原论文下载)MUCNetV2:内存瓶颈和计算负载问题一举突破?分类&检测都有较高性能(附源代码下载)旋转角度目标检测的重要性!!!(附源论文下载)双尺度残差检测器:无先验检测框进行目标检测(附论文下载)Fast YOLO:用于实时嵌入式目标检测(附论文下载)Micro-YOLO:探索目标检测压缩模型的有效方法(附论文下载)目标检测干货 | 多级特征重复使用大幅度提升检测精度(文末附论文下载)发布于 2021-12-07 22:51算法深度学习(Deep Learning)多模态学习赞同 9添加评论分享喜欢收藏申请转载文章被以下专栏收录计算机视觉研究院一起学习计算机视觉领域知识,共同
Nat Genet︱浙江大学医学院郭国骥/韩晓平团队发表基于人工智能神经网络的基因组解读系统Nvwa,并揭示细胞命运决定的共性规律 - 知乎
Nat Genet︱浙江大学医学院郭国骥/韩晓平团队发表基于人工智能神经网络的基因组解读系统Nvwa,并揭示细胞命运决定的共性规律 - 知乎首发于逻辑神经科学(中文阅读)切换模式写文章登录/注册Nat Genet︱浙江大学医学院郭国骥/韩晓平团队发表基于人工智能神经网络的基因组解读系统Nvwa,并揭示细胞命运决定的共性规律逻辑神经科学以逻辑之学术思维,探索神经科学之奥秘来源︱“逻辑神经科学”姊妹号“岚翰生命科学”撰文︱郭国骥,李佳琦,王晶晶责编︱王思珍,方以一编辑︱夏 叶,王如华预测基因表达和解析基因调控机制一直是基因组学的重要目标。尽管研究人员已经努力使用细胞系或组织中的各种实验特征来预测调节信号和基因表达[1-3],但在单细胞分辨率下进行生物体规模的表达预测仍然具有挑战性。如今单细胞图谱能够以统一的标准呈现物种细胞的表型[4-9],因而人类有机会使用跨物种的单细胞数据来探索进化过程中不同细胞类型的表达和调控程序。研究团队假设可以直接从基因组序列预测生物体规模的单细胞基因表达,并试图在具有巨大细胞类型多样性的后生动物中检验这一假设。2022年10月13日,浙江大学基础医学院/浙江省良渚实验室郭国骥教授/韩晓平教授团队在Nature Genetics上发表了题为“Deep learning of cross-species single cell landscapes identifies conserved regulatory programs underlying cell types”的研究。该研究利用自主构建的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型保守的调控程序。构建了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞分辨率下的基因表达。斑马鱼、果蝇和蚯蚓作为后生动物重要的进化节点,全身单细胞转录组图谱的绘制将有助于解析物种进化进程中细胞命运的决定机制。研究人员首先使用其团队自主研发的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图(图1a)。其中,斑马鱼图谱收集了635228个单细胞数据,果蝇图谱涵盖了276706个单细胞数据,蚯蚓图谱包含了95,020个单细胞数据。共计定义了105个斑马鱼细胞类型和1285个细胞亚型,87个果蝇类型和1085个细胞亚类,以及62个蚯蚓细胞类型和462个细胞亚类。该研究利用这三种模式动物的单细胞图谱,并结合其他五种代表性动物的单细胞图谱(人类[4]、小鼠[5]、海鞘[10]、线虫[11]和涡虫[12]),挖掘了跨物种细胞谱系特异性的转录因子,探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型(图1b),特别是免疫细胞、基质细胞、神经元、上皮细胞、内皮细胞和生殖细胞的保守调节程序(图1c)。图1 斑马鱼、果蝇和蚯蚓的单细胞转录图谱的构建和跨物种分析(图源:Jiaqi Li, et al., Nat Genet, 2022)基于DNA序列编码基因表达模式的假设,研究人员提出了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞水平的基因表达,且预测准确度与实验测量精度相当(图2a)。值得注意的是,Nvwa模型可以高度准确地预测几乎所有测试物种的基因表达,并且保持物种细胞图谱所描绘的细胞类型特异性。除了预测基因表达,研究人员应用模型预测整个基因组的转录调控信号,模型预测与功能基因组学数据描绘的调控区域高度一致。此后,接着研究团队通过解释模型的预测能力,来分析其生物学意义,由此揭示模型识别的具有可预测性的调控模式。通过检查模型第一层卷积的序列特征Filter,团队揭示了细胞类型特异的基序,其中部分基序与转录因子结合基序具有一致性。并且在特异细胞类型中的调控模式作用与细胞类型特异性富集的转录因子基序相一致(图2b,c)。基于Nvwa模型Filter的跨物种比较,该研究还发现同源Filter倾向于保持跨物种的细胞类型特异性。该工作首次建立了物种层面基因组编码细胞图谱的整合模型,并为解码多物种基因调控程序提供了宝贵资源。图2 深度学习模型Nvwa(女娲)(图源:Jiaqi Li, et al., Nat Genet, 2022)文章结论与讨论,启发与展望综上所述,该研究利用自主构建的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型保守的调控程序。该研究基于单细胞图谱提出了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞分辨率下的基因表达。该研究基于Nvwa模型学习衍生的谱系特异性基序,表征了跨物种细胞类型特异性的调节程序。值得一提的是,Nvwa模型将为组学和精准医疗研究提供强大的技术支撑。例如:基于Nvwa模型可以实现多组学大数据整合的序列建模;利用Nvwa模型可以大规模解码单细胞尺度下的疾病/肿瘤基因组,进一步开发基因组学功能预测工具;应用Nvwa模型解析基因组序列的特性,实现DNA序列突变效应的预测,有助于筛选与复杂疾病关联的突变效应等。尽管Nvwa为研究进化过程中细胞类型特异性调控程序提供了一个新的视角,但Nvwa模型的解释和应用仍然需要谨慎。首先,超参数和模型结构,特别是第一层卷积Filter,需要权衡序列模式解释的简并性和灵敏度。其次,在本研究中,研究人员将深度学习衍生序列模式解释为转录因子的基序,存在一些新序列模式无法分配到已知数据库。另外,通过功能实验验证调节元件是非常重要的。最后,基因调控机制是复杂的,模型的结构、预测性能和调控逻辑解释仍需改进。原文链接:https://www.nature.com/articles/s41588-022-01197-7通讯作者:郭国骥教授(左),韩晓平教授(中),王晶晶研究员(右)(图源:照片提供自郭国骥/韩晓平团队)作者简介郭国骥,浙江大学医学院教授,浙江省良渚实验室核心PI,博士生导师,浙江大学医学院干细胞与再生医学中心副主任,浙江大学血液学研究所副所长,浙江大学干细胞联盟副主席。2017年获“国家优秀青年基金”,2019年入选“万人计划”科技创新领军人才。曾获“树兰医学青年奖”,“霍英东青年教师奖”,“细胞生物学会青年科学家奖”等荣誉。一直致力于单细胞分析技术的开发与应用,并在细胞图谱的绘制上有突出贡献;在Nature, Cell, Nature Genetics, Cell Stem Cell等著名期刊发表多篇学术论文。韩晓平,理学博士,浙江大学医学院教授,博士生导师,国家优秀青年科学基金获得者,浙江大学求是青年学者。主要从事单细胞分析技术方向的研究,以第一作者或通讯作者在Nature, Cell,Cell Research,Nature Genetics等顶级期刊发表多篇研究论文。王晶晶,生物信息学博士,浙江省良渚实验室特聘研究员,主要从事单细胞组学大数据整合,细胞类型进化等研究,以第一作者或共同第一作者在Nature,Nature Genetics,Cell Reports等杂志发表多篇论文。往期文章精选【1】J Neuroinflammation︱卓业鸿/苏文如团队揭示铁死亡可能是视网膜缺血-再灌注的损伤新机制和治疗新靶点【2】PNAS︱钟毅课题组揭示Rac1-依赖的遗忘机制是情绪状态影响记忆表达的神经基础【3】Cell Metab 综述︱曹旭团队评述骨内感知系统调控骨稳态及骨痛【4】NAN︱苑林宏课题组揭示DHA干预对ApoE-/-和C57 WT小鼠脑脂质水平、脂肪酸转运体表达和Aβ代谢有不同影响【5】Autophagy 综述︱李晓江团队评述线粒体自噬在体内和体外模型中的差异及研究进展【6】HBM︱商慧芳课题组通过功能影像技术揭示帕金森病的运动进展标记物【7】Cell Rep︱宋建人课题组揭示脊髓损伤后脊髓环路重建的新规律【8】HBM︱宋艳/孙黎课题组基于机器学习技术揭示首发ADHD儿童内隐视觉空间编码障碍的认知神经基【9】Nat Neurosci︱突破!冷泉港实验室李波课题组揭示泛杏仁核结构调控饮食选择与能量代谢的神经机制【10】Nat Commun︱邢大军课题组揭示微眼跳方向特异性调制视觉信息编码新机制参考文献1.Agarwal V, Shendure J. Predicting mRNA Abundance Directly from Genomic Sequence Using Deep Convolutional Neural Networks. Cell Rep. 2020, 31(7):107663.2.Zhou J, Troyanskaya OG. Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015, 12(10):931-4.3.Kelley DR, Snoek J, Rinn JL. Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 2016, 26(7):990-9.4.Han X, Zhou Z, Fei L, Sun H, Wang R, Chen Y, Chen H, Wang J, Tang H, Ge W, Zhou Y, Ye F, Jiang M, Wu J, Xiao Y, Jia X, Zhang T, Ma X, Zhang Q, Bai X, Lai S, Yu C, Zhu L, Lin R, Gao Y, Wang M, Wu Y, Zhang J, Zhan R, Zhu S, Hu H, Wang C, Chen M, Huang H, Liang T, Chen J, Wang W, Zhang D, Guo G. Construction of a human cell landscape at single-cell level. Nature. 2020, 581(7808):303-309.5.Han X, Wang R, Zhou Y, Fei L, Sun H, Lai S, Saadatpour A, Zhou Z, Chen H, Ye F, Huang D, Xu Y, Huang W, Jiang M, Jiang X, Mao J, Chen Y, Lu C, Xie J, Fang Q, Wang Y, Yue R, Li T, Huang H, Orkin SH, Yuan GC, Chen M, Guo G. Mapping the Mouse Cell Atlas by Microwell-Seq. Cell. 2018, 172(5):1091-1107.e17.6.Fei L, Chen H, Ma L, E W, Wang R, Fang X, Zhou Z, Sun H, Wang J, Jiang M, Wang X, Yu C, Mei Y, Jia D, Zhang T, Han X, Guo G. Systematic identification of cell-fate regulatory programs using a single-cell atlas of mouse development. Nat Genet. 2022, 54(7):1051-1061.7.Ye F, Zhang G, E W, Chen H, Yu C, Yang L, Fu Y, Li J, Fu S, Sun Z, Fei L, Guo Q, Wang J, Xiao Y, Wang X, Zhang P, Ma L, Ge D, Xu S, Caballero-Pérez J, Cruz-Ramírez A, Zhou Y, Chen M, Fei JF, Han X, Guo G. Construction of the axolotl cell landscape using combinatorial hybridization sequencing at single-cell resolution. Nat Commun. 2022, 13(1):4228.8.Liao Y, Ma L, Guo Q, E W, Fang X, Yang L, Ruan F, Wang J, Zhang P, Sun Z, Chen H, Lin Z, Wang X, Wang X, Sun H, Fang X, Zhou Y, Chen M, Shen W, Guo G, Han X. Cell landscape of larval and adult Xenopus laevis at single-cell resolution. Nat Commun. 2022, 13(1):4306.9.Wang R, Zhang P, Wang J, Ma L, E W, Suo S, Jiang M, Li J, Chen H, Sun H, Fei L, Zhou Z, Zhou Y, Chen Y, Zhang W, Wang X, Mei Y, Sun Z, Yu C, Shao J, Fu Y, Xiao Y, Ye F, Fang X, Wu H, Guo Q, Fang X, Li X, Gao X, Wang D, Xu PF, Zeng R, Xu G, Zhu L, Wang L, Qu J, Zhang D, Ouyang H, Huang H, Chen M, Ng SC, Liu GH, Yuan GC, Guo G, Han X. Construction of a cross-species cell landscape at single-cell level. Nucleic Acids Res. 2022, gkac633.10.Cao C, Lemaire LA, Wang W, Yoon PH, Choi YA, Parsons LR, Matese JC, Wang W, Levine M, Chen K. Comprehensive single-cell transcriptome lineages of a proto-vertebrate. Nature. 2019, 571(7765):349-354.11.Cao J, Packer JS, Ramani V, Cusanovich DA, Huynh C, Daza R, Qiu X, Lee C, Furlan SN, Steemers FJ, Adey A, Waterston RH, Trapnell C, Shendure J. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science. 2017, 357(6352):661-667.12.Fincher CT, Wurtzel O, de Hoog T, Kravarik KM, Reddien PW. Cell type transcriptome atlas for the planarian Schmidtea mediterranea. Science. 2018, 360(6391):eaaq1736.本文完发布于 2022-11-29 16:18・IP 属地河南浙江大学医学院浙江大学人工智能赞同 2添加评论分享喜欢收藏申请转载文章被以下专栏收录逻辑神经科学(中文阅读)以严谨的学术逻辑思维,探索神经科学的奥秘岚翰生命科学(中文阅读)关爱生命科学,报道最新进展;立思维,学知识,
Home | NVWA-English
Home | NVWA-English
Go to content
You are here:
Home
Search within English part of NVWA-English
Search
Highlights
Approved Establishments
Import Regulations of the Netherlands on Plant Health
Brexit
Bringing pets into the Netherlands
Travelling to and from the United Kingdom with your pet after Brexit
About us
Frequently Asked Questions about Babboe cargo bikes (15-02-2024)
Image: ©NVWA
Main menu
Animal health
Keep animals healthy. Animal welfare. Animal products processing secure
Plant health
Plant diseases and prevent pests.
Food safety
Supervising the production, preparation, transport and sale of food products and sale of tobacco products.
Product safety
Safe personal care products, indoor and outdoor games, consumer products in and around the house.
Home
News
Bird flu (Avian influenza) now as well in Croatia: mandatory additional Cleansing & Disinfection transport vehicles in The Netherlands
Image: ©NVWA
01-02-2024 | 16:19
Due to the outbreak of contagious (highly pathogenic) bird flu (HPAI) at kept poultry in Croatia transport vehicles for poultry or hatching eggs that return or enter from Croatia to the Netherlands require additional cleansing and disinfection (C&D) immediately after entering the Netherlands.
Bird flu (Avian influenza) now as well in Sweden: mandatory additional Cleansing & Disinfection transport vehicles in The Netherlands
Image: ©NVWA
18-01-2024 | 15:06
Due to the outbreak of contagious (highly pathogenic) bird flu (HPAI) at kept poultry in Sweden transport vehicles for poultry or hatching eggs that return or enter from Sweden to The Netherlands require additional cleansing and disinfection (C&D) immediately after entering the Netherlands.
Bird flu (Avian influenza) now as well in Germany and Belgium, mandatory additional Cleansing & Disinfection transport vehicles in The Netherlands
Image: ©NVWA
15-12-2023 | 08:51
Due to the outbreak of contagious (highly pathogenic) bird flu (HPAI) at kept poultry in Germany and Belgium transport vehicles for poultry [1] or hatching eggs that return or enter from Germany and Belgium to The Netherlands require additional cleansing and disinfection (C&D) immediately after entering The Netherlands.
More news
The NVWA, because we stand for the safety of food and consumer products, animal welfare and nature.
Service
Contact
RSS
Help
Documents
About this site
Copyright
Privacy
Cookies
Accessibility
Report vulnerability
This website in other languages:
Nederlands
Deep learning of cross-species single-cell landscapes identifies conserved regulatory programs underlying cell types | Nature Genetics
Deep learning of cross-species single-cell landscapes identifies conserved regulatory programs underlying cell types | Nature Genetics
Skip to main content
Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Advertisement
View all journals
Search
Log in
Explore content
About the journal
Publish with us
Subscribe
Sign up for alerts
RSS feed
nature
nature genetics
articles
article
Article
Published: 13 October 2022
Deep learning of cross-species single-cell landscapes identifies conserved regulatory programs underlying cell types
Jiaqi Li1,2 na1, Jingjing Wang
ORCID: orcid.org/0000-0002-6006-27271,2 na1, Peijing Zhang1,2 na1, Renying Wang1 na1, Yuqing Mei1, Zhongyi Sun1, Lijiang Fei1, Mengmeng Jiang1,2, Lifeng Ma1, Weigao E1, Haide Chen1,2, Xinru Wang1, Yuting Fu1, Hanyu Wu1, Daiyuan Liu1, Xueyi Wang1, Jingyu Li
ORCID: orcid.org/0000-0001-9677-75321, Qile Guo3, Yuan Liao1,4, Chengxuan Yu1, Danmei Jia1, Jian Wu5, Shibo He
ORCID: orcid.org/0000-0002-1505-67666, Huanju Liu
ORCID: orcid.org/0000-0002-3370-42967, Jun Ma
ORCID: orcid.org/0000-0002-1609-32947, Kai Lei8, Jiming Chen
ORCID: orcid.org/0000-0003-3155-31456, Xiaoping Han
ORCID: orcid.org/0000-0003-3201-76351,4 & …Guoji Guo
ORCID: orcid.org/0000-0002-1716-46211,2,3,4,9 Show authors
Nature Genetics
volume 54, pages 1711–1720 (2022)Cite this article
21k Accesses
9 Citations
112 Altmetric
Metrics details
Subjects
Cell biologyComputational biology and bioinformatics
AbstractDespite extensive efforts to generate and analyze reference genomes, genetic models to predict gene regulation and cell fate decisions are lacking for most species. Here, we generated whole-body single-cell transcriptomic landscapes of zebrafish, Drosophila and earthworm. We then integrated cell landscapes from eight representative metazoan species to study gene regulation across evolution. Using these uniformly constructed cross-species landscapes, we developed a deep-learning-based strategy, Nvwa, to predict gene expression and identify regulatory sequences at the single-cell level. We systematically compared cell-type-specific transcription factors to reveal conserved genetic regulation in vertebrates and invertebrates. Our work provides a valuable resource and offers a new strategy for studying regulatory grammar in diverse biological systems.
Access through your institution
Buy or subscribe
This is a preview of subscription content, access via your institution
Access options
Access through your institution
Access through your institution
Change institution
Buy or subscribe
Access Nature and 54 other Nature Portfolio journalsGet Nature+, our best-value online-access subscription24,99 € / 30 dayscancel any timeLearn moreSubscribe to this journalReceive 12 print issues and online access195,33 € per yearonly 16,28 € per issueLearn moreRent or buy this articlePrices vary by article typefrom$1.95to$39.95Learn morePrices may be subject to local taxes which are calculated during checkout
Additional access options:
Log in
Learn about institutional subscriptions
Read our FAQs
Contact customer support
Fig. 1: Zebrafish and Drosophila cell landscapes constructed using Microwell-seq.Fig. 2: Cross-species analysis of eight species.Fig. 3: Comparison of regulatory TFs within and across species using scRNA-seq data.Fig. 4: Application and interpretation of the deep-learning model framework.Fig. 5: Interpretation of the deep-learning model framework.
Data availability
Raw data files for the RNA sequencing analysis reported in this paper have been deposited in is the NCBI Gene Expression Omnibus under accession number GSE178151. Digital expression matrices are available at https://figshare.com/s/ecc05b1051fb5678fd3e. Nvwa data can be accessed at http://bis.zju.edu.cn/nvwa/.
Code availability
The source code for reproducing our analysis and running and training the Nvwa models is available at GitHub (https://github.com/JiaqiLiZju/Nvwa/) and Zenodo (https://zenodo.org/record/6806748) (JiaqiLiZju/Nvwa: release v.1.0, 2022).
ReferencesHan, X. et al. Construction of a human cell landscape at single-cell level. Nature 581, 303–309 (2020).Article
CAS
PubMed
Google Scholar
Tabula Sapiens, C. et al. The Tabula Sapiens: a multiple-organ, single-cell transcriptomic atlas of humans. Science 376, eabl4896 (2022).Article
Google Scholar
Han, X. et al. Mapping the Mouse Cell Atlas by Microwell-Seq. Cell 173, 1307 (2018).Article
CAS
PubMed
Google Scholar
Tabula Muris, C. et al. Single-cell transcriptomics of 20 mouse organs creates a Tabula Muris. Nature 562, 367–372 (2018).Article
Google Scholar
Jiang, M. M. et al. Characterization of the zebrafish cell landscape at single-cell resolution. Front. Cell Dev. Biol. 9, 743421 (2021).Article
PubMed
PubMed Central
Google Scholar
Cao, J. Y. et al. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science 357, 661–667 (2017).Article
CAS
PubMed
PubMed Central
Google Scholar
Cao, C. et al. Comprehensive single-cell transcriptome lineages of a proto-vertebrate. Nature 571, 349–354 (2019).Article
CAS
PubMed
PubMed Central
Google Scholar
Fincher, C. T., Wurtzel, O., de Hoog, T., Kravarik, K. M. & Reddien, P. W. Cell type transcriptome atlas for the planarian Schmidtea mediterranea. Science 360, eaaq1736 (2018).Article
PubMed
PubMed Central
Google Scholar
Sebe-Pedros, A. et al. Cnidarian cell type diversity and regulation revealed by whole-organism single-cell RNA-seq. Cell 173, 1520–1534.e20 (2018).Article
CAS
PubMed
Google Scholar
Li, H. et al. Fly Cell Atlas: A single-nucleus transcriptomic atlas of the adult fruit fly. Science 375, eabk2432 (2022).Article
CAS
PubMed
PubMed Central
Google Scholar
Arendt, D. et al. The origin and evolution of cell types. Nat. Rev. Genet. 17, 744–757 (2016).Article
CAS
PubMed
Google Scholar
Wang, J. et al. Tracing cell-type evolution by cross-species comparison of cell atlases. Cell Rep. 34, 108803 (2021).Article
CAS
PubMed
Google Scholar
Agarwal, V. & Shendure, J. Predicting mRNA abundance directly from genomic sequence using deep convolutional neural networks. Cell Rep. 31, 107663 (2020).Article
CAS
PubMed
Google Scholar
Zhou, J. & Troyanskaya, O. G. Predicting effects of noncoding variants with deep learning-based sequence model. Nat. Methods 12, 931–934 (2015).Article
CAS
PubMed
PubMed Central
Google Scholar
Kelley, D. R., Snoek, J. & Rinn, J. L. Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 26, 990–999 (2016).Article
CAS
PubMed
PubMed Central
Google Scholar
Avsec, Z. et al. Effective gene expression prediction from sequence by integrating long-range interactions. Nat. Methods 18, 1196–1203 (2021).Article
CAS
PubMed
PubMed Central
Google Scholar
Maslova, A. et al. Deep learning of immune cell differentiation. Proc. Natl Acad. Sci. USA 117, 25655–25666 (2020).Article
CAS
PubMed
PubMed Central
Google Scholar
Satija, R., Farrell, J. A., Gennert, D., Schier, A. F. & Regev, A. Spatial reconstruction of single-cell gene expression data. Nat. Biotechnol. 33, 495–502 (2015).Article
CAS
PubMed
PubMed Central
Google Scholar
Buchon, N., Silverman, N. & Cherry, S. Immunity in Drosophila melanogaster–from microbial recognition to whole-organism physiology. Nat. Rev. Immunol. 14, 796–810 (2014).Article
CAS
PubMed
PubMed Central
Google Scholar
Krausgruber, T. et al. Structural cells are key regulators of organ-specific immune responses. Nature 583, 296–302 (2020).Article
CAS
PubMed
PubMed Central
Google Scholar
van Dijk, D. et al. Recovering gene interactions from single-cell data using data diffusion. Cell 174, 716 (2018).Article
PubMed
PubMed Central
Google Scholar
Tarashansky, A. J. et al. Mapping single-cell atlases throughout Metazoa unravels cell type evolution. eLife 10, e66747 (2021).Article
CAS
PubMed
PubMed Central
Google Scholar
Rui, L., Schmitz, R., Ceribelli, M. & Staudt, L. M. Malignant pirates of the immune system. Nat. Immunol. 12, 933–940 (2011).Article
CAS
PubMed
Google Scholar
Zhou, J. et al. Deep learning sequence-based ab initio prediction of variant effects on expression and disease risk. Nat. Genet. 50, 1171–1179 (2018).Article
CAS
PubMed
PubMed Central
Google Scholar
Kelley, D. R. et al. Sequential regulatory activity prediction across chromosomes with convolutional neural networks. Genome Res. 28, 739–750 (2018).Article
CAS
PubMed
PubMed Central
Google Scholar
de Souza, N. The ENCODE project. Nat. Methods 9, 1046 (2012).Article
PubMed
Google Scholar
Srivastava, A. K. & Schlessinger, D. Structure and organization of ribosomal DNA. Biochimie 73, 631–638 (1991).Article
CAS
PubMed
Google Scholar
Suzuki, H., Moriwaki, K. & Sakurai, S. Sequences and evolutionary analysis of mouse 5S rDNAs. Mol. Biol. Evol. 11, 704–710 (1994).CAS
PubMed
Google Scholar
Zentner, G. E., Balow, S. A. & Scacheri, P. C. Genomic characterization of the mouse ribosomal DNA locus. G3 4, 243–254 (2014).Article
CAS
PubMed
Google Scholar
Aibar, S. et al. SCENIC: single-cell regulatory network inference and clustering. Nat. Methods 14, 1083–1086 (2017).Article
CAS
PubMed
PubMed Central
Google Scholar
Gupta, S., Stamatoyannopoulos, J. A., Bailey, T. L. & Noble, W. S. Quantifying similarity between motifs. Genome Biol. 8, R24 (2007).Article
PubMed
PubMed Central
Google Scholar
Cusanovich, D. A. et al. A single-cell atlas of in vivo mammalian chromatin accessibility. Cell 174, 1309–1324.e18 (2018).Article
CAS
PubMed
PubMed Central
Google Scholar
Hannenhalli, S. & Kaestner, K. H. The evolution of Fox genes and their role in development and disease. Nat. Rev. Genet. 10, 233–240 (2009).Article
CAS
PubMed
PubMed Central
Google Scholar
Shafer, M. E. R. Cross-species analysis of single-cell transcriptomic data. Front. Cell Dev. Biol. 7, 175 (2019).Article
PubMed
PubMed Central
Google Scholar
Picelli, S. et al. Full-length RNA-seq from single cells using Smart-seq2. Nat. Protoc. 9, 171–181 (2014).Article
CAS
PubMed
Google Scholar
Dobin, A. et al. STAR: ultrafast universal RNA-seq aligner. Bioinformatics 29, 15–21 (2013).Article
CAS
PubMed
Google Scholar
Zheng, G. X. et al. Massively parallel digital transcriptional profiling of single cells. Nat. Commun. 8, 14049 (2017).Article
CAS
PubMed
PubMed Central
Google Scholar
McGinnis, C. S., Murrow, L. M. & Gartner, Z. J. DoubletFinder: doublet detection in single-cell RNA sequencing data using artificial nearest neighbors. Cell Syst. 8, 329–337.e4 (2019).Article
CAS
PubMed
PubMed Central
Google Scholar
Wolf, F. A., Angerer, P. & Theis, F. J. SCANPY: large-scale single-cell gene expression data analysis. Genome Biol. 19, 15 (2018).Article
PubMed
PubMed Central
Google Scholar
Stuart, T. et al. Comprehensive integration of single-cell data. Cell 177, 1888–1902.e21 (2019).Article
CAS
PubMed
PubMed Central
Google Scholar
Shao, Y. et al. Genome and single-cell RNA-sequencing of the earthworm Eisenia andrei identifies cellular mechanisms underlying regeneration. Nat. Commun. 11, 2656 (2020).Article
CAS
PubMed
PubMed Central
Google Scholar
Rozanski, A. et al. PlanMine 3.0–improvements to a mineable resource of flatworm biology and biodiversity. Nucleic Acids Res. 47, D812–D820 (2019).Article
CAS
PubMed
Google Scholar
Satou, Y., Kawashima, T., Shoguchi, E., Nakayama, A. & Satoh, N. An integrated database of the ascidian, Ciona intestinalis: towards functional genomics. Zool. Sci. 22, 837–843 (2005).Article
CAS
Google Scholar
Haas, B. J. et al. De novo transcript sequence reconstruction from RNA-seq using the Trinity platform for reference generation and analysis. Nat. Protoc. 8, 1494–1512 (2013).Article
CAS
PubMed
Google Scholar
Emms, D. M. & Kelly, S. OrthoFinder: solving fundamental biases in whole genome comparisons dramatically improves orthogroup inference accuracy. Genome Biol. 16, 157 (2015).Article
PubMed
PubMed Central
Google Scholar
Crow, M., Paul, A., Ballouz, S., Huang, Z. J. & Gillis, J. Characterizing the replicability of cell types defined by single cell RNA-sequencing data using MetaNeighbor. Nat. Commun. 9, 884 (2018).Article
PubMed
PubMed Central
Google Scholar
Fischer, S., Crow, M., Harris, B. D. & Gillis, J. Scaling up reproducible research for single-cell transcriptomics using MetaNeighbor. Nat. Protoc. 16, 4031–4067 (2021).Article
CAS
PubMed
PubMed Central
Google Scholar
Hu, H. et al. AnimalTFDB 3.0: a comprehensive resource for annotation and prediction of animal transcription factors. Nucleic Acids Res. 47, D33–D38 (2019).Article
CAS
PubMed
Google Scholar
dos Santos, G. et al. FlyBase: introduction of the Drosophila melanogaster Release 6 reference genome assembly and large-scale migration of genome annotations. Nucleic Acids Res. 43, D690–D697 (2015).Article
PubMed
Google Scholar
Dubaj Price, M. & Hurd, D. D. WormBase: a model organism database. Med. Ref. Serv. Q. 38, 70–80 (2019).Article
PubMed
Google Scholar
Li, H. et al. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25, 2078–2079 (2009).Article
PubMed
PubMed Central
Google Scholar
Grant, C. E., Bailey, T. L. & Noble, W. S. FIMO: scanning for occurrences of a given motif. Bioinformatics 27, 1017–1018 (2011).Article
CAS
PubMed
PubMed Central
Google Scholar
Fornes, O. et al. JASPAR 2020: update of the open-access database of transcription factor binding profiles. Nucleic Acids Res. 48, D87–D92 (2020).CAS
PubMed
Google Scholar
Harris, T. W. et al. WormBase: a modern model organism information resource. Nucleic Acids Res. 48, D762–D767 (2020).CAS
PubMed
Google Scholar
Zhao, H. et al. CrossMap: a versatile tool for coordinate conversion between genome assemblies. Bioinformatics 30, 1006–1007 (2014).Article
PubMed
Google Scholar
Ramirez, F., Dundar, F., Diehl, S., Gruning, B. A. & Manke, T. deepTools: a flexible platform for exploring deep-sequencing data. Nucleic Acids Res. 42, W187–W191 (2014).Article
CAS
PubMed
PubMed Central
Google Scholar
Yu, G., Wang, L. G. & He, Q. Y. ChIPseeker: an R/Bioconductor package for ChIP peak annotation, comparison and visualization. Bioinformatics 31, 2382–2383 (2015).Article
CAS
PubMed
Google Scholar
Ramirez, F. et al. High-resolution TADs reveal DNA sequences underlying genome organization in flies. Nat. Commun. 9, 189 (2018).Article
PubMed
PubMed Central
Google Scholar
Simonyan, K., Vedaldi, A. & Zisserman, A. Deep inside convolutional networks: visualising image classification models and saliency maps. Preprint at https://doi.org/10.48550/arXiv.1312.6034 (2014).Weirauch, M. T. et al. Determination and inference of eukaryotic transcription factor sequence specificity. Cell 158, 1431–1443 (2014).Article
CAS
PubMed
PubMed Central
Google Scholar
Download referencesAcknowledgementsG.G. is a participant in the Human Cell Atlas Project. We thank M. Chen, Y. Zhou, F. Gu, D. Wang, P. Xu, C. Li, K. Li and H. Wu for support on the project. We thank M20 (Hangzhou), G-BIO (Hangzhou), BGI (Shenzhen) and CNGB (Shenzhen) for supporting the sequencing experiments; and Vazyme (Nanjing) for supplying the customized enzymes used in the study. We also thank the core facility platform of Zhejiang University School of Medicine and the Center of Cryo-Electron Microscopy at Zhejiang University for computational resources, and the core facilities of Zhejiang University Medical Center and the Liangzhu Laboratory for technical support. This work was supported by National Natural Science Foundation of China grants 31930028 to G.G., 31922049 to X.H., 91842301 to G.G., 32000461 to J.W. and 62088101 to J.C.; National Key Research and Development Program grants 2018YFA0800503 to G.G., 2018YFA0107804 to G.G. and 2018YFA0107801 to X.H.; Fundamental Research Funds for the Central Universities (G.G.); and Alibaba-Zhejiang University Joint Research Center of Future Digital Healthcare.Author informationAuthor notesThese authors contributed equally: Jiaqi Li, Jingjing Wang, Peijing Zhang, Renying Wang.Authors and AffiliationsCenter for Stem Cell and Regenerative Medicine and Bone Marrow Transplantation Center of the First Affiliated Hospital, Zhejiang University School of Medicine, Hangzhou, ChinaJiaqi Li, Jingjing Wang, Peijing Zhang, Renying Wang, Yuqing Mei, Zhongyi Sun, Lijiang Fei, Mengmeng Jiang, Lifeng Ma, Weigao E, Haide Chen, Xinru Wang, Yuting Fu, Hanyu Wu, Daiyuan Liu, Xueyi Wang, Jingyu Li, Yuan Liao, Chengxuan Yu, Danmei Jia, Xiaoping Han & Guoji GuoLiangzhu Laboratory, Zhejiang University Medical Center, Hangzhou, ChinaJiaqi Li, Jingjing Wang, Peijing Zhang, Mengmeng Jiang, Haide Chen & Guoji GuoZhejiang University-University of Edinburgh Institute, Zhejiang University School of Medicine, Zhejiang University, Hangzhou, ChinaQile Guo & Guoji GuoZhejiang Provincial Key Laboratory for Tissue Engineering and Regenerative Medicine, Dr. Li Dak Sum & Yip Yio Chin Center for Stem Cell and Regenerative Medicine, Hangzhou, ChinaYuan Liao, Xiaoping Han & Guoji GuoDivision of Hepatobiliary and Pancreatic Surgery, Department of Surgery, First Affiliated Hospital School of Medicine, Zhejiang University, Hangzhou, ChinaJian WuCollege of Control Science and Engineering, Zhejiang University, Hangzhou, ChinaShibo He & Jiming ChenWomen’s Hospital and Institute of Genetics, Zhenjiang University School of Medicine, Hangzhou, ChinaHuanju Liu & Jun MaWestlake Laboratory of Life Sciences and Biomedicine, Key Laboratory of Growth Regulation and Translational Research of Zhejiang Province, School of Life Sciences, Westlake University, Hangzhou, ChinaKai LeiAlibaba-Zhejiang University Joint Research Center of Future Digital Healthcare, Hangzhou, ChinaGuoji GuoAuthorsJiaqi LiView author publicationsYou can also search for this author in
PubMed Google ScholarJingjing WangView author publicationsYou can also search for this author in
PubMed Google ScholarPeijing ZhangView author publicationsYou can also search for this author in
PubMed Google ScholarRenying WangView author publicationsYou can also search for this author in
PubMed Google ScholarYuqing MeiView author publicationsYou can also search for this author in
PubMed Google ScholarZhongyi SunView author publicationsYou can also search for this author in
PubMed Google ScholarLijiang FeiView author publicationsYou can also search for this author in
PubMed Google ScholarMengmeng JiangView author publicationsYou can also search for this author in
PubMed Google ScholarLifeng MaView author publicationsYou can also search for this author in
PubMed Google ScholarWeigao EView author publicationsYou can also search for this author in
PubMed Google ScholarHaide ChenView author publicationsYou can also search for this author in
PubMed Google ScholarXinru WangView author publicationsYou can also search for this author in
PubMed Google ScholarYuting FuView author publicationsYou can also search for this author in
PubMed Google ScholarHanyu WuView author publicationsYou can also search for this author in
PubMed Google ScholarDaiyuan LiuView author publicationsYou can also search for this author in
PubMed Google ScholarXueyi WangView author publicationsYou can also search for this author in
PubMed Google ScholarJingyu LiView author publicationsYou can also search for this author in
PubMed Google ScholarQile GuoView author publicationsYou can also search for this author in
PubMed Google ScholarYuan LiaoView author publicationsYou can also search for this author in
PubMed Google ScholarChengxuan YuView author publicationsYou can also search for this author in
PubMed Google ScholarDanmei JiaView author publicationsYou can also search for this author in
PubMed Google ScholarJian WuView author publicationsYou can also search for this author in
PubMed Google ScholarShibo HeView author publicationsYou can also search for this author in
PubMed Google ScholarHuanju LiuView author publicationsYou can also search for this author in
PubMed Google ScholarJun MaView author publicationsYou can also search for this author in
PubMed Google ScholarKai LeiView author publicationsYou can also search for this author in
PubMed Google ScholarJiming ChenView author publicationsYou can also search for this author in
PubMed Google ScholarXiaoping HanView author publicationsYou can also search for this author in
PubMed Google ScholarGuoji GuoView author publicationsYou can also search for this author in
PubMed Google ScholarContributionsG.G., X.H. and J. Wang conceived the study. G.G. and X.H. supervised the study. Jiaqi Li designed the model. X.H., R.W., M.J., X.H., H.C., Xinru Wang, Xueyi Wang, Y.L., D.J. and T.Z. performed all the experiments. Jiaqi Li, J. Wang, P.Z., Y.M., Z.S., L.F., L.M., W.E., Y.F., H.W., D.L., H.W., Jingyu Li, Q.G. and C.Y. performed all computational analyses. H.L., J.M. and K.L. helped with the cell-type annotation. J. Wu., S.H. and J.C. guided model design and parameter optimization. G.G., J. Wang., Jiaqi Li and P.Z. wrote the initial draft of the manuscript. All authors participated in discussion of results and manuscript editing.Corresponding authorsCorrespondence to
Jingjing Wang, Xiaoping Han or Guoji Guo.Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Genetics thanks the anonymous reviewers for their contribution to the peer review of this work.
Additional informationPublisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.Extended dataExtended Data Fig. 1 Zebrafish cell landscape was constructed using Microwell-seq.(a-b) t-SNE visualization of 635,228 single cells from whole bodies across three stages of zebrafish, colored by stage (a) and cell lineage (b). (c) Heatmap showing the scaled average expression levels of zebrafish cell type-specific marker genes (left), and relative gene expression of representative cell type-specific markers for each cell type overlaid on t-SNE plots (right). (d) Heatmap showing the correspondence between zebrafish cell landscape in this study (row) and tissue-specific zebrafish dataset from Jiang et al., 2021 (column). Blue refers to a mean AUROC greater than 0.9. (e-f) t-SNE showing 24 subclusters from zebrafish C1 (neuron) (e), and violin plot showing the scale normalized expression levels of representative markers for each subcluster (f).Extended Data Fig. 2 Drosophila cell landscape was constructed using Microwell-seq.(a-b) t-SNE visualization of 276,706 single cells from whole bodies across two stages of Drosophila, colored by stage (a) and cell lineage (b). (c) Heatmap showing the scaled average expression levels of Drosophila cell type-specific marker genes (left), and relative gene expression of representative cell type-specific markers for each cluster overlaid on t-SNE plots (right). (d) Heatmap showing the correspondence between Drosophila cell landscape in this study (row) and tissue-specific fly cell atlas construed by Li et al., 2021 (column). Blue refers to a mean AUROC greater than 0.9. (e-f) t-SNE showing 21 subclusters from Drosophila C1 (central nerve cell) (e), violin plot showing the scale normalized expression levels of representative markers for each subcluster (f).Extended Data Fig. 3 Earthworm cell landscape was constructed using Microwell-seq.(a) t-SNE visualization of 95,020 single cells from whole bodies of earthworm, colored by cell type (left) and cell lineage (right). (b) Heatmap showing the scaled average expression levels of earthworm cell type-specific marker genes (left), and relative gene expression of representative cell type-specific markers for each cluster overlaid on t-SNE plots (right). (c) The hierarchical clustering tree (right) showing the similarity among earthworm 62 cells, and the histogram plot (left) showing the subtypes of each cell type. The similarity refers to the AUROC score from MetaNeighbor analysis. (d-e) t-SNE showing 11 subclusters from earthworm C24 (neuron) (d), violin plot showing the scale normalized expression levels of representative markers for each subcluster (e).Extended Data Fig. 4 Characteristics of immune-related structure cells in zebrafish, Drosophila, and earthworm.(a) Ridge plots showing the scale normalized expression levels of known zebrafish epithelial markers and immune-related markers in zebrafish epithelial cells. (b) Ridge plots showing the the scale normalized expression levels of known Drosophila MT markers and immune-related markers in Drosophila MT cells (C34 and C69). (c) Ridge plots showing the scale normalized expression levels of two earthworm immune-related markers, evm.TU.Chr04.2499 and evm.TU.ctg2984.2 in earthworm coelomocytes (earthworm C0, C5, C6, C16, C19, C30, C32, C35, and C37). (d-e) t-SNE visualization of 11 subtypes of MT cells from Drosophila C34 and C69, and violin plot showing the scale normalized expression levels of representative markers for each subtype (e).Extended Data Fig. 5 Cross-species analysis among eight species.(a) Sankey diagrams showing homologous cell-type pairs between human and mouse obtained from SAMap analyses based on different datasets. (b) The Venn diagram showing the number of overlapping homologous cell-type pairs obtained from SAMap analyses based on different datasets. (c-e) The information of homologous cell types among human, mouse, zebrafish, Ciona, Drosophila, earthworm, C. elegans, and planarian, including the number of homologous cell-type pairs (c), aligned score (d), and number of enriched gene pairs (e). Boxplots represent the median, 25th percentile, and 75th percentile, and whiskers correspond to 1.5 times the interquartile range. N of cell types: human: 434, mouse: 432, zebrafish: 378, Ciona: 291, Drosophila: 288, earthworm: 110, C. elegans: 321, and planarians: 173.Extended Data Fig. 6 Screening conserved regulators underlying cell lineages.(a) The bar chart showing the percentage of cell lineage-specific TFs in humans, mice, zebrafish, Ciona, Drosophila, earthworm, C. elegans, and planarians across different conservative levels based on homologous genes (left) obtained from SAMap and 1-to-1 orthologous genes (right). The group number refers to the conservative level. TFs were divided into eight different conservative levels (Level 1–8) based on the conversion of homologous TFs among eight species. Level 1 means that TFs have no homologous genes in other species, and Level 8 means that TFs have homologous genes in all other seven species. (b) The bar chart showing the conserved neuron-related TFs between human and other species. The blue bar refers to the number of homologous TF pairs between human and other species, the yellow bar refers to the number of human neuron-related TFs involved in homologous gene pairs, and the grey bar refers to the number of other species neuron-related TFs involved in homologous gene pairs. (c) Conserved neuron-related regulators across eight species showing a comprehensive homologous link from lower to higher organisms. (d) Sankey diagram showing homologous relationships among vertebrates’ immune-related TFs. (e) Sankey diagram showing homologous relationships of immune-related TFs between Drosophila and human.Extended Data Fig. 7 The performance of model training for eight species.(a) The AUROC values of major cell lineages for humans, mice, zebrafish, Ciona, Drosophila, earthworm, C. elegans, and planarians. (b) Boxplots of AUPR values for eight species. Boxplots represent the median, 25th percentile, and 75th percentile, and whiskers correspond to 1.5 times the interquartile range. N of cells: humans: 134,557, mice: 179,344, zebrafish: 241,233, Ciona: 12,489, Drosophila: 77,337, earthworm: 29,609, C. elegans: 30,515, and planarians: 50,562.Extended Data Fig. 8 Benchmark of prediction performance.(a) The AUROC values of Nvwa, Basset, DeepSEA, Beluga, Basenji, SVM, random labels and random features on human (n = 134,557) and Drosophila (n = 77,337) specific datasets. (b) The AUROC of multiple genome training for zebrafish (n = 241,233) and C. elegans (n = 30,515). Boxplots represent the median, 25th percentile, and 75th percentile, and whiskers correspond to 1.5 times the interquartile range. (c) The heatmaps showing the correlation between observed and Nvwa-predicted cell type-specific transcription for eight species.Extended Data Fig. 9 Overview of sequence patterns recognized by Nvwa.(a) Mean saliency scores show transcriptional start site and important information-rich region recognized by Nvwa. (b) Comparison of the first-layer convolution filters derived from feature map-based approaches and gradient-based TF-MoDISco on Drosophila-specific model. (c) Examples of known TFBS compared with the PWMs of Nvwa first-layer in humans, mice, zebrafish, Ciona, Drosophila, C. elegans, and planarians.Extended Data Fig. 10 Overview of sequence motifs and their influence recognized by Nvwa.(a) Volcano plot of Nvwa first-layer filters for humans, mice, zebrafish, Ciona, Drosophila, earthworm, C. elegans, and planarians. The x-axis represents the information contents (IC) of a Filter, the y-axis represents the overall influence on of a Filter, Filters with high influence are tagged as up, and Filters with low influence are tagged as down. Those reductant Filters are tagged as triangle and non-reductant Filters are tagged as dots, the size of elements represents the reproducibility in each independent cross-validation run. (b) Barplot of the Nvwa and single-cell ATAC cell type specific motifs for mouse. Neuronal cells (C12, n=29 and C5, n=169 for Nvwa and sci-ATAC data respectively) and endothelial cells (C50, n=31 and C22, n=136 for Nvwa and sci-ATAC data respectively) were shown. Hit indicates the same motifs, NotHit indicates the different motifs identified by TomTom. X-axis indicates the percentage. (c) Barplot of the Nvwa and single -cell ATAC cell type specific transcription factor for Drosophila. Neuronal cells (C1, n=293, CB, n=639 and OL, n=484 for Nvwa and Flybrain (GSE163697) data respectively) were shown. Hit indicates the same TFs, and NotHit indicates the different TFs annotated by TomTom. X-axis indicates the percentage.Supplementary informationSupplementary InformationSupplementary Figures 1–7.Reporting SummarySupplementary Tables 1–36.Rights and permissionsSpringer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.Reprints and permissionsAbout this articleCite this articleLi, J., Wang, J., Zhang, P. et al. Deep learning of cross-species single-cell landscapes identifies conserved regulatory programs underlying cell types.
Nat Genet 54, 1711–1720 (2022). https://doi.org/10.1038/s41588-022-01197-7Download citationReceived: 18 July 2021Accepted: 31 August 2022Published: 13 October 2022Issue Date: November 2022DOI: https://doi.org/10.1038/s41588-022-01197-7Share this articleAnyone you share the following link with will be able to read this content:Get shareable linkSorry, a shareable link is not currently available for this article.Copy to clipboard
Provided by the Springer Nature SharedIt content-sharing initiative
This article is cited by
Delineating mouse β-cell identity during lifetime and in diabetes with a single cell atlas
Karin HrovatinAimée Bastidas-PonceFabian J. Theis
Nature Metabolism (2023)
Inflammation and aging: signaling pathways and intervention therapies
Xia LiChentao LiHe Huang
Signal Transduction and Targeted Therapy (2023)
Convergent differentiation of multiciliated cells
Shinhyeok ChaeTae Joo ParkTaejoon Kwon
Scientific Reports (2023)
Access through your institution
Buy or subscribe
Access through your institution
Change institution
Buy or subscribe
Associated content
Deciphering single-cell transcriptional programs across species
Nature Genetics
Research Briefing
20 Oct 2022
Advertisement
Explore content
Research articles
Reviews & Analysis
News & Comment
Current issue
Collections
Follow us on Facebook
Follow us on Twitter
Subscribe
Sign up for alerts
RSS feed
About the journal
Aims & Scope
Journal Information
Journal Metrics
Our publishing models
Editorial Values Statement
Editorial Policies
Content Types
About the Editors
Web Feeds
Posters
Contact
Research Cross-Journal Editorial Team
Reviews Cross-Journal Editorial Team
Publish with us
Submission Guidelines
For Reviewers
Language editing services
Submit manuscript
Search
Search articles by subject, keyword or author
Show results from
All journals
This journal
Search
Advanced search
Quick links
Explore articles by subject
Find a job
Guide to authors
Editorial policies
Nature Genetics (Nat Genet)
ISSN 1546-1718 (online)
ISSN 1061-4036 (print)
nature.com sitemap
About Nature Portfolio
About us
Press releases
Press office
Contact us
Discover content
Journals A-Z
Articles by subject
Protocol Exchange
Nature Index
Publishing policies
Nature portfolio policies
Open access
Author & Researcher services
Reprints & permissions
Research data
Language editing
Scientific editing
Nature Masterclasses
Research Solutions
Libraries & institutions
Librarian service & tools
Librarian portal
Open research
Recommend to library
Advertising & partnerships
Advertising
Partnerships & Services
Media kits
Branded
content
Professional development
Nature Careers
Nature
Conferences
Regional websites
Nature Africa
Nature China
Nature India
Nature Italy
Nature Japan
Nature Korea
Nature Middle East
Privacy
Policy
Use
of cookies
Your privacy choices/Manage cookies
Legal
notice
Accessibility
statement
Terms & Conditions
Your US state privacy rights
© 2024 Springer Nature Limited
Close banner
Close
Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.
Email address
Sign up
I agree my information will be processed in accordance with the Nature and Springer Nature Limited Privacy Policy.
Close banner
Close
Get the most important science stories of the day, free in your inbox.
Sign up for Nature Briefing
Mythology: 女娲补天 – Nvwa Mends the Heavens – Chinese Reading Practice
Mythology: 女娲补天 – Nvwa Mends the Heavens – Chinese Reading Practice
Skip to the content
Search
Chinese Reading PracticeSimplified Chinese Reading Exercises with English Translations
Menu
Beginner
Intermediate
Advanced
By topic
Biography
Children’s Stories
Essays
Idioms
History
Jokes
Myths & Fables
News
Novels
Poetry
Politics & Communism
About
Search
Search for:
Close search
Close Menu
Beginner
Intermediate
Advanced
By topicShow sub menu
Biography
Children’s Stories
Essays
Idioms
History
Jokes
Myths & Fables
News
Novels
Poetry
Politics & Communism
About
Categories
Advanced
Mythology: 女娲补天 – Nvwa Mends the Heavens
Doesn’t it seem like humankind has a collective memory of some prehistoric natural disaster built into our DNA? So many cultures have ancient myths about close calls with total destruction, be it Noah’s Ark, or the story of Atlantis, or in China’s case, “Nvwa Mends the Heavens” (女娲补天).
Post author
By Kendra
Post date
May 21, 2020
1 Comment on Mythology: 女娲补天 – Nvwa Mends the Heavens
Doesn’t it seem like humankind has a collective memory of some prehistoric natural disaster built into our DNA? So many cultures have ancient myths about close calls with total destruction, be it Noah’s Ark, or the story of Atlantis, or in China’s case, “Nvwa Mends the Heavens” (女娲补天).
This ancient myth is old, very old. The goddess Nvwa is not really part of any modern Buddhist or Taoist pantheon, she appears to come from shamanistic folk legends that predate these religions. Early versions of this myth were included in texts from the Western Han (206 BC – 9 AD), so the story of Nvwa has been floating around for at least 2000 years. There are two main myths regarding Nvwa, our post today is one of them. In the other one, she creates human beings out of mud and clay.
Some language stuff
As always, I’ve highlighted the proper nouns. We’ve got the water god Gonggong (共工), the fire god Zhurong (祝融), Buzhou Mountain (不周山), and the heroine of our tale, the goddess Nvwa (女娲).
I didn’t see anything in here that might trip up advanced readers, probably just new vocab, but there was one weird thing: the placement of the character 晨 seems a little weird to me. It means “morning”, as in 早晨, but I’m only 80% sure I translated its usage in this context correctly. If you have more insight, let me know in the comments.
女娲补天
传说当人类繁衍起来后,忽然水神共工和火神祝融打起仗来,他们从天上一直打到地下,闹得到处不宁,结果祝融打胜了,但败了的共工不服,一怒之下,把头撞向不周山。
不周山崩裂了,撑支天地之间的大柱断折了,天倒下了半边,出现了一个大窟窿,地也陷成一道道大裂纹,山林烧起了大火,洪水从地底下喷涌出来,龙蛇猛兽也出来吞食人民。
人类面临着空前大灾难。女娲目睹人类遭到如此奇祸,感到无比痛苦,于是决心补天,以终止这场灾难。她选用各种各样的五色石子,架起火将它们熔化成浆,用这种石浆将残缺的天窟窿填好,随后又斩下一只大龟的四脚,当作四根柱子把倒塌的半边天支起来。
女娲还擒杀了残害人民的黑龙,刹住了龙蛇的嚣张气焰。最后为了堵住洪水不再漫流,女娲还收集了大量芦草,把它们烧成灰,埋塞向四处铺开的洪流。
经过女娲一番辛劳整治,苍天总算补上了,地填平了,水止住了,龙蛇猛兽佥欠迹了,人民又重新过着安乐的生活。但是这场特大的灾祸毕竟留下了痕迹。从此天还是有些向西北倾斜,因此太阳、月亮和众星晨都很自然地归向西方,又因为地向东南倾斜,所以一切江河都往那里汇流。
Show English translation »
Legend has it that when humans had just begun to multiply [upon the earth], a great battle suddenly broke out between the water god Gonggong and the fire god Zhurong, disrupting tranquility in every quarter. In the end, Zhurong was victorious, but Gonggong would not accept his defeat, and in a fit of anger, smashed his head against Buzhou Mountain.
Buzhou Mountain split apart, the giant pillars that held up the heavens were broken, and half of the sky caved in, leaving a large hole. Great cracks also appeared in the earth, the mountain forests went up in flame, floodwaters spurted forth from under the ground, and dragons and monsters came out and began devouring the people.
Mankind was facing an unprecedented catastrophe. When Nvwa witnessed the unexpected disaster that befell humanity, she was terribly agonized, and so she decided to patch up the sky, in order to put an end to the present calamity. She chose to use a selection of multi-colored stones, set a great fire to smelt them into a slurry, and used the smelted rock to paste over the hole in the firmament. Then she chopped the four feet off a giant tortoise, and used them as the four pillars to prop up half the heavens.
Nvwa also captured and slayed a great black dragon that was cruelly slaughtering human beings, putting a stop to the dragon’s arrogance. Finally, in order to stem the floodwaters, Nvwa collected many reeds and grasses, burned them into ash, and stopped up the streams that were spreading in all directions.
After Nvwa had toiled through the repairs, the sky was mended, the earth was even, the waters had been stopped, no trace remained of the dragons and monsters, and the people once again lived in peace and happiness. But of course, some vestiges of this huge calamity remained. From then on, the sky was slightly tilted to the northwest, and because of this the sun, moon and stars all naturally return to the west each morning, and also because the earth is tilted to the southeast, all of the rivers flow in that direction.
Tags
Myths & Fables
←
Children’s Story: 有学问的儿子 – The Learned Son
→
Children’s Story: 最可口的食物 – The most appetizing meal
1 reply on “Mythology: 女娲补天 – Nvwa Mends the Heavens”
Heathersays:
July 17, 2020 at 12:27 am
It’s actually 星’辰’, a fancy way to say ‘stars’. 北极星, for example, is also called 北辰. The character 辰 itself generally has to do with all things celestial/related to time and space. The ancient unit for measuring time, for example, is called 时辰 (you’ll see it mentioned a lot in historical drama)! ^^
Reply
Leave a Reply Cancel replyYour email address will not be published. Required fields are marked *Comment Name *
Email *
Website
Save my name, email, and website in this browser for the next time I comment.
Δ
Whose fault is this website?Kendra Schaefer is a tech researcher based in Beijing since 2002. This is her pet project.
Twitter: @ChineseReading
©
2024 Chinese Reading Practice
Powered by WordPress
To the top ↑
Up ↑
“女娲”来了!良渚实验室郭国骥团队Nature Genetics发文 提出深度学习模型Nvwa并揭示细胞命运决定的共性规律 - 科学研究 - 良渚实验室
“女娲”来了!良渚实验室郭国骥团队Nature Genetics发文 提出深度学习模型Nvwa并揭示细胞命运决定的共性规律 - 科学研究 - 良渚实验室
首页
实验室概况
实验室简介
组织架构
图库
公共平台
平台概况
技术平台
通知公告
下载专区
联系我们
研究领域
人才队伍
首席科学家
课题组长
人才招聘
合作研发
医学创新基金
政校企合作
社会捐赠
党建群团
党建之家
信息公开
班车信息
表格下载
物资采购
内部信息
内部通知
公示
主任办公会纪要
文件制度
电话黄页
首页
实验室概况
实验室简介
组织架构
图库
公共平台
平台概况
技术平台
通知公告
下载专区
联系我们
研究领域
人才队伍
首席科学家
课题组长
人才招聘
合作研发
医学创新基金
政校企合作
社会捐赠
党建群团
党建之家
信息公开
班车信息
表格下载
物资采购
内部信息
内部通知
公示
主任办公会纪要
文件制度
电话黄页
科学研究
科学研究
“女娲”来了!良渚实验室郭国骥团队Nature Genetics发文 提出深度学习模型Nvwa并揭示细胞命运决定的共性规律
发布日期:2022-10-14
点击次数:
2022年10月13日晚,Nature Genetics在线刊登了良渚实验室郭国骥教授和浙江大学基础医学院韩晓平教授团队的学术论文“Deep learning of cross-species single cell landscapes identifies conserved regulatory programs underlying cell types”。该研究利用自主构建的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型保守的调控程序。此外,该研究提出了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞分辨率下的基因表达。该研究基于Nvwa模型学习衍生的谱系特异性基序,表征了跨物种细胞类型特异性的调节程序。 预测基因表达和解析基因调控机制一直是基因组学的重要目标。在单细胞分辨率下进行生物体规模的表达预测对研究人员而言仍具有挑战性。如今单细胞图谱可以统一的标准呈现物种细胞的表型,因此有机会使用跨物种的单细胞数据来探索进化过程中不同细胞类型的表达和调控程序。研究团队假设可以直接从基因组序列预测生物体规模的单细胞基因表达,并试图在具有巨大细胞类型多样性的后生动物中检验这一假设。 该研究中,研究人员首先使用其团队自主研发的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图。其中,斑马鱼图谱收集了635,228个单细胞数据,果蝇图谱涵盖了276,706个单细胞数据,蚯蚓图谱包含了95,020个单细胞数据。该研究利用这三种模式动物的单细胞图谱,并结合其他五种代表性动物的单细胞图谱(人类、小鼠、海鞘、线虫和涡虫),挖掘了跨物种细胞谱系特异性的转录因子,探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型,特别是免疫细胞、基质细胞、神经元、上皮细胞、内皮细胞和生殖细胞的保守调节程序。 研究团队表示,斑马鱼、果蝇和蚯蚓作为后生动物重要的进化节点,全身单细胞转录组图谱的绘制将有助于解析物种进化进程中细胞命运的决定机制。基于DNA序列编码基因表达模式的假设,该研究提出了深度学习模型Nvwa,首次实现了完全基于基因组序列预测单细胞水平的基因表达,且预测准确度与实验测量精度相当。值得注意的是,Nvwa模型可以高度准确地预测几乎所有测试物种的基因表达。此外,通过检查模型第一层的卷积的基序特征Filter,团队揭示了细胞类型特异的基序。这些基序与在特异细胞类型中作用机制明确的转录因子基序相一致。基于Nvwa模型Filter的跨物种比较,该研究还发现同源Filter倾向于保持跨物种的细胞类型特异性。该工作首次建立了物种层面基因组编码细胞图谱的整合模型,并为解码多物种基因调控程序提供了宝贵资源。 良渚实验室(系统医学与精准诊治浙江省实验室)坚持“四个面向”,着重创新思维模式将临床治疗和基础医学研究连接起来,强调“从临床到基础”以及“从基础到临床”的双向转化,更好地服务人民生命健康。值得一提的是,Nvwa模型将为组学和精准医疗研究提供强大的技术支撑。例如:基于Nvwa模型可以实现多组学大数据整合的序列建模;利用Nvwa模型可以大规模解码单细胞尺度下的疾病/肿瘤基因组,进一步开发基因组学功能预测工具;应用Nvwa模型解析基因组序列的特性,实现DNA序列突变效应的预测,有助于筛选与复杂疾病关联的突变效应等。 良渚实验室特聘研究员王晶晶、浙江大学基础医学院2019级直博生李佳琦、博士后张霈婧和汪仁英为本文共同第一作者。郭国骥教授、韩晓平教授和王晶晶研究员为本文的通讯作者。该研究获得了国家重点研发计划和国家自然科学基金的支持。 原文链接:https://www.nature.com/articles/s41588-022-01197-7
Copyright @ 2021 良渚实验室. 版权所有
技术支持:
创高软件
管理入口
地址 : 浙江省杭州市余杭区文一西路1369号
电话 : 0571-88790516
传真 : 0571-88790516
Email: liangzhu@zju.edu.cn
友情链接
浙江大学
浙江大学
浙江大学医学中心
良渚实验室管理系统
欢迎关注良渚实验室公众号
【学术前沿】郭国骥/韩晓平团队开发基于人工智能神经网络的基因组解读系统——女娲,并揭示细胞命运决定…_澎湃号·政务_澎湃新闻-The Paper
沿】郭国骥/韩晓平团队开发基于人工智能神经网络的基因组解读系统——女娲,并揭示细胞命运决定…_澎湃号·政务_澎湃新闻-The Paper下载客户端登录无障碍+1【学术前沿】郭国骥/韩晓平团队开发基于人工智能神经网络的基因组解读系统——女娲,并揭示细胞命运决定…2022-10-14 17:00来源:澎湃新闻·澎湃号·政务字号预测基因表达和解析基因调控机制一直是基因组学的重要目标。尽管研究人员已经努力使用细胞系或组织中的各种实验特征来预测调节信号和基因表达【1-3】,但在单细胞分辨率下进行生物体规模的表达预测仍然具有挑战性。如今单细胞图谱能够以统一的标准呈现物种细胞的表型【4-9】,因而人类有机会使用跨物种的单细胞数据来探索进化过程中不同细胞类型的表达和调控程序。2022年10月13日,浙江大学基础医学院/浙江省良渚实验室郭国骥/韩晓平团队在Nature Genetics上发表了文章 Deep learning of cross-species single cell landscapes identifies conserved regulatory programs underlying cell types 。该研究利用自主构建的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图谱,并探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型保守的调控程序。此外,该研究提出了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞分辨率下的基因表达。该研究基于Nvwa模型学习衍生的谱系特异性基序,表征了跨物种细胞类型特异性的调节程序。研究团队假设可以直接从基因组序列预测生物体规模的单细胞基因表达,并试图在具有巨大细胞类型多样性的后生动物中检验这一假设。该研究中,研究人员首先使用其团队自主研发的高通量单细胞测序平台Microwell-seq绘制了斑马鱼、果蝇和蚯蚓的全身单细胞转录组图。其中,斑马鱼图谱收集了635,228个单细胞数据,果蝇图谱涵盖了276,706个单细胞数据,蚯蚓图谱包含了95,020个单细胞数据。该研究利用这三种模式动物的单细胞图谱,并结合其他五种代表性动物的单细胞图谱(人类【5】、小鼠【5】、海鞘【10】、线虫【11】和涡虫【12】),挖掘了跨物种细胞谱系特异性的转录因子,探究了八种代表性后生动物细胞类型的跨物种可比性,揭示了脊椎动物细胞类型,特别是免疫细胞、基质细胞、神经元、上皮细胞、内皮细胞和生殖细胞的保守调节程序。基于DNA序列编码基因表达模式的假设,该研究提出了深度学习模型Nvwa(女娲),首次实现了完全基于基因组序列预测单细胞水平的基因表达,且预测准确度与实验测量精度相当。值得注意的是,Nvwa模型可以高度准确地预测几乎所有测试物种的基因表达。此外,通过检查模型第一层的卷积的基序特征Filter,团队揭示了细胞类型特异的基序。这些基序与在特异细胞类型中作用机制明确的转录因子基序相一致。基于模Nvwa模型Filter的跨物种比较,该研究还发现同源Filter倾向于保持跨物种的细胞类型特异性。该工作首次建立了物种层面基因组编码细胞图谱的整合模型,并为解码多物种基因调控程序提供了宝贵资源。浙江大学基础医学院2019级直博生李佳琦、良渚实验室特聘研究员王晶晶、浙江大学基础医学院博士后张霈婧和汪仁英为本文共同第一作者。浙江大学基础医学院郭国骥教授、韩晓平教授和良渚实验室王晶晶研究员为本文的通讯作者。原文链接:https://doi.org/10.1038/s41588-022-01197-7参考文献1. Agarwal V, Shendure J. Predicting mRNA Abundance Directly from Genomic Sequence Using Deep Convolutional Neural Networks. Cell Rep. 2020, 31(7):107663.2. Zhou J, Troyanskaya OG. Predicting effects of noncoding variants with deep learning-based sequence model. Nat Methods. 2015, 12(10):931-4.3. Kelley DR, Snoek J, Rinn JL. Basset: learning the regulatory code of the accessible genome with deep convolutional neural networks. Genome Res. 2016, 26(7):990-9.4. Han X, Zhou Z, Fei L, Sun H, Wang R, Chen Y, Chen H, Wang J, Tang H, Ge W, Zhou Y, Ye F, Jiang M, Wu J, Xiao Y, Jia X, Zhang T, Ma X, Zhang Q, Bai X, Lai S, Yu C, Zhu L, Lin R, Gao Y, Wang M, Wu Y, Zhang J, Zhan R, Zhu S, Hu H, Wang C, Chen M, Huang H, Liang T, Chen J, Wang W, Zhang D, Guo G. Construction of a human cell landscape at single-cell level. Nature. 2020, 581(7808):303-309.5. Han X, Wang R, Zhou Y, Fei L, Sun H, Lai S, Saadatpour A, Zhou Z, Chen H, Ye F, Huang D, Xu Y, Huang W, Jiang M, Jiang X, Mao J, Chen Y, Lu C, Xie J, Fang Q, Wang Y, Yue R, Li T, Huang H, Orkin SH, Yuan GC, Chen M, Guo G. Mapping the Mouse Cell Atlas by Microwell-Seq. Cell. 2018, 172(5):1091-1107.e17.6. Fei L, Chen H, Ma L, E W, Wang R, Fang X, Zhou Z, Sun H, Wang J, Jiang M, Wang X, Yu C, Mei Y, Jia D, Zhang T, Han X, Guo G. Systematic identification of cell-fate regulatory programs using a single-cell atlas of mouse development. Nat Genet. 2022, 54(7):1051-1061.7. Ye F, Zhang G, E W, Chen H, Yu C, Yang L, Fu Y, Li J, Fu S, Sun Z, Fei L, Guo Q, Wang J, Xiao Y, Wang X, Zhang P, Ma L, Ge D, Xu S, Caballero-Pérez J, Cruz-Ramírez A, Zhou Y, Chen M, Fei JF, Han X, Guo G. Construction of the axolotl cell landscape using combinatorial hybridization sequencing at single-cell resolution. Nat Commun. 2022, 13(1):4228.8. Liao Y, Ma L, Guo Q, E W, Fang X, Yang L, Ruan F, Wang J, Zhang P, Sun Z, Chen H, Lin Z, Wang X, Wang X, Sun H, Fang X, Zhou Y, Chen M, Shen W, Guo G, Han X. Cell landscape of larval and adult Xenopus laevis at single-cell resolution. Nat Commun. 2022, 13(1):4306.9. Wang R, Zhang P, Wang J, Ma L, E W, Suo S, Jiang M, Li J, Chen H, Sun H, Fei L, Zhou Z, Zhou Y, Chen Y, Zhang W, Wang X, Mei Y, Sun Z, Yu C, Shao J, Fu Y, Xiao Y, Ye F, Fang X, Wu H, Guo Q, Fang X, Li X, Gao X, Wang D, Xu PF, Zeng R, Xu G, Zhu L, Wang L, Qu J, Zhang D, Ouyang H, Huang H, Chen M, Ng SC, Liu GH, Yuan GC, Guo G, Han X. Construction of a cross-species cell landscape at single-cell level. Nucleic Acids Res. 2022, gkac633.10. Cao C, Lemaire LA, Wang W, Yoon PH, Choi YA, Parsons LR, Matese JC, Wang W, Levine M, Chen K. Comprehensive single-cell transcriptome lineages of a proto-vertebrate. Nature. 2019, 571(7765):349-354.11. Cao J, Packer JS, Ramani V, Cusanovich DA, Huynh C, Daza R, Qiu X, Lee C, Furlan SN, Steemers FJ, Adey A, Waterston RH, Trapnell C, Shendure J. Comprehensive single-cell transcriptional profiling of a multicellular organism. Science. 2017, 357(6352):661-667.12. Fincher CT, Wurtzel O, de Hoog T, Kravarik KM, Reddien PW. Cell type transcriptome atlas for the planarian Schmidtea mediterranea. Science. 2018, 360(6391):eaaq1736.(可上下滑动阅览)原标题:《【学术前沿】郭国骥/韩晓平团队开发基于人工智能神经网络的基因组解读系统——女娲,并揭示细胞命运决定的共性规律》阅读原文特别声明本文为澎湃号作者或机构在澎湃新闻上传并发布,仅代表该作者或机构观点,不代表澎湃新闻的观点或立场,澎湃新闻仅提供信息发布平台。申请澎湃号请用电脑访问http://renzheng.thepaper.cn。+1收藏我要举报查看更多查看更多开始答题扫码下载澎湃新闻客户端Android版iPhone版iPad版关于澎湃加入澎湃联系我们广告合作法律声明隐私政策澎湃矩阵澎湃新闻微博澎湃新闻公众号澎湃新闻抖音号IP SHANGHAISIXTH TONE新闻报料报料热线: 021-962866报料邮箱: news@thepaper.cn沪ICP备14003370号沪公网安备31010602000299号互联网新闻信息服务许可证:31120170006增值电信业务经营许可证:沪B2-2017116© 2014-2024 上海东方报业有限公“NVWA”陈士华:区块链+乡村振兴,NFT可以是很好的媒介_腾讯新闻
“NVWA”陈士华:区块链+乡村振兴,NFT可以是很好的媒介_腾讯新闻
“NVWA”陈士华:区块链+乡村振兴,NFT可以是很好的媒介
本次销售花园村数字村民NFT所有资金将全部直接进入冢头镇联合社,用于村庄发展。
链新(ID:ChinaBlockchainNews)原创
作者 | 冯铭
12月5日,全球首个连接实体经济的数字作品电商平台NVWA正式上线。首批上线的产品是基于冢头镇花园村的数字村民NFT权益项目,限量发行200份。本次销售花园村数字村民NFT所有资金将全部直接进入冢头镇联合社,用于村庄发展。
NVWA为湖南简通科技有限公司开发的数字作品交易平台,旨在以区块链底层技术的数字作品赋能实体经济,推动数字经济与实体经济的高度融合。简通科技十分重视合规运营,严格遵守国家相关法律法规。
发行数字村民NFT的初衷与目的是什么?在可预见的未来,NFT可以为“三农”带来哪些新的生机和活力?NFT未来监管趋势又会怎样?对此,《链新》专访了NVWA平台项目合作负责人陈士华。
陈士华,参与乡村建设16年,一直在实践中追逐理想,梦想在国内实践蒙德拉贡合作社。曾先后创办多个农业项目,从2018年开始探索区块链在实体经济领域的应用。
1
链新
带领村民奔赴共同富裕
《链新》:你参与发起花园村数字村民NFT项目有哪些契机和原因?
陈士华:冢头镇在河南省平顶山市郏县,当地49个村,每个村都成立了村集体经济股份合作社,同时49个村的集体合作社发起成立镇级联合社,这说明当地政府非常具有公信力,愿意带领村民一起共同富裕。这非常符合乡村振兴的精神,以及当前倡导的共同富裕思想,以村民为发展主体,大力发展乡村集体经济。同时冢头镇还有很多古村落及文化遗产,这些都是当地乡村振兴得天独厚的资源。
我们团队中有几名核心成员都是从事十多年乡村建设的志愿者,和冢头镇一直有密切联系。我们一直认为科技发展最终都应该推动实体经济的发展,这是基础,所以我们也一直在探索区块链技术在乡村振兴中能够发挥的推动作用,而NFT可以是很好的媒介,就像电子邮箱之于互联网,NFT可能在区块链网络中也扮演类似的角色。此外,乡村生态资源的价值化,也需要一个媒介与外部市场和资源产生化学反应。正是在这种契机下,我们和当地相关部门深入探讨区块链和乡村振兴的结合,顺势推出花园村数字村民NFT。
《链新》:如何诠释数字村民这个概念?
陈士华:数字村民类似于荣誉村民或新村民,乡村资产归属于村集体,由村民共同所有,所以村民这个身份背后是有财产关系的。非本村人是无法获得村民身份,户口从农村迁出后,再迁回来也是禁止的。这种权属关系有好的一面,保护了村民的权益,但同时也阻碍了资金、资源和人才的流入。
数字村民希望创造一种新的村民身份,一方面要为村里整合各种资源,包括资金和人才,参与当地村庄建设;另一方面数字村民还能根据对村庄发展的贡献,享受当地村集体的部分权益和分红。
《链新》:NFT可以为“三农”带来哪些新的生机和活力?
陈士华:NFT是方便实现乡村生态资源价值化的方式之一,NFT可以大大降低非标准、不可分割资产的交易成本和流通成本。乡村生态资源在完成一级市场定价后,村集体经济组织(或镇联社或县平台公司)可以在区块链上发行本村生态资源使用权和经营权的NFT,这些NFT可以在专门的NFT平台进行交易,如此可以不但让村民获得可持续的财产性收益,还能通过资金和人才的流入发展振兴当地产业,这也是乡村振兴战略中新提出的实现共同富裕的方式。
2
链新
数字作品赋能实体经济
《链新》:NVWA的定位是全球首个连接实体经济的数字作品电商平台。从治理层面看,如何解决链上数字作品和链下资产脱钩的问题?
陈士华:NVWA定位首家赋能实体经济的数字作品电商平台,以实体资产为基础发行的数字作品是NVWA平台的主要产品。发行方都将通过NVWA平台的严格选择,大家对价值认同的一致使我们非常看重的,也是建立信任、长远合作发展的基础。最初选取信用程度比较高的企业或组织,比如村集体、国企等。我们首发项目的合作方是冢头镇政府和冢头镇级联合社。若当前持有数字作品的人需要交割,可以按照发行方设置的规则与发行方进行线下交割。
《链新》:对于创作者,如何保护作品的版权?投资者又该如何鉴别作品的真伪?
陈士华:数字作品是在区块链上发行的数字资产,具有不可篡改、唯一性、不可分割、可追溯等特点,数字作品的创作者拥有其在区块链上唯一的数字代码,这就可以保护其版权。交易后,购买者就拥有其唯一的数字代码。数字作品的所有交易都可以在区块链浏览器上进行查询,可以明确知道数字作品属于谁。在数字作品交易市场上,只有拥有者才能将其售卖。对于NVWA数字作品平台上的很多产品,版权和真伪的判断就更容易了,都是与实体资产绑定的,那些资产就在那里。
《链新》:目前,业界对数字作品的定义仍有争议,相关交易平台也面临监管挑战。在你看来,交易平台跟现在的电商平台本质上有哪些区别?数字作品未来监管趋势会怎样?
陈士华:数字作品交易本质是资产交易,电商平台本质是产品交易,数字作品本身具有投资属性,不是以销售额或利润最大化为目的。绝大部分数字作品数量都有限制。数字作品平台的主要风险在于若被用于投机炒作,将影响金融安全。我们希望政府能尽快出台相关监管政策,在政策范围内开展数字作品交易才是长久之计。
《链新》:在花园村数字村民NFT项目推进过程中,你觉得最大的难点是什么?
陈士华:现在地方政府对区块链和新兴科技还是非常欢迎和接受,政府也非常希望能够为当地发展提供助力,前期沟通合作过程中基本没有太多困难。如果一定要说有难点,应该还是任何一个企业都需要面临的项目运营本身。随着产品的上线,运营不断推进的过程中,困难和问题会越来越多,所以我们也做好了各种预案,充实内部团队,积极向外学习。
在花园村,村子里有几栋特别漂亮的、有历史感的老屋,过去村民因为一些历史原因始终没能在边界问题上达成一致,导致了某种程度上的废弃。花园村的现任支书上任后,很快了解清楚前因后果,协调完毕村民间的矛盾,把这几栋老屋利用起来,还修葺一番,预备作为村子发展起来未来对外接待的民宿。不止于此,他还是利用“历史遗产”的高手:废弃的烤烟房被改造成现代化的公共厕所,难以利用的荒地改造成停车场、废弃的电线杆再次利用作为村内游乐设施的支柱。
他还整合村内的资源,预备利用独特的交通和文化优势,来发展研学旅游产业。村里一直养牛嘛,那就增加玩法,做小牛农场,城里的孩子们可以来认养,周末来玩;集体的菜地,种上各种各样的菜,一块块分割开来,也计划分租给城里的家庭,便于休闲度假时携家带口而来,连过去农村常用的板车,都手工做成了迷你版本,供小朋友拉。而这一切,都是他带领村委会和村民们自行探索出来的。
这也解答了为什么我们会首先选择做花园村数字村民NFT项目。过去看了太多案例,村庄的发展和规划完全交给外来的投资方,而外来投资方不了解村里的历史文化地理环境,做出来的项目不仅水土不服,村民也没法参与进来,浪费很大的资源。而花园村是非常少见的本村人完全参与,能够最大限度挖掘利用本村优势的发展道路。当然,出于长远发展的考虑,村子也需要更多来自外界的刺激,包括资源人才等方面。
本文为链新(ID:ChinaBlockchainNews)原创,未经授权禁止擅自转载。