Spatial Join Techniques
- 格式:pdf
- 大小:557.89 KB
- 文档页数:71
为了成为一名画家我应该做什么英语作文Becoming a painter is a journey filled with passion, dedication, and a relentless pursuit of artistic expression. As someone aspiring to join the ranks of renowned artists, there are several key steps you can take to pave the way for a successful career in painting.Firstly, it is essential to cultivate a deep and genuine love for art. Painting is not merely a skill to be acquired; it is a means of conveying emotions, ideas, and perspectives that resonate with the viewer. Immerse yourself in the world of art by visiting art galleries, museums, and exhibitions. Observe the works of masters, study their techniques, and try to understand the underlying messages and emotions they convey. This exposure will help you develop a critical eye and a deeper appreciation for the art form.Secondly, hone your technical skills through consistent practice and experimentation. Painting is a craft that requires patience, discipline, and a willingness to learn. Engage in regular drawing and sketching exercises to improve your hand-eye coordination, spatial awareness, and ability to translate your vision onto the canvas. Experiment withdifferent mediums, such as oils, acrylics, watercolors, and pastels, to find the one that best suits your artistic style and preferences.As you progress, it is essential to develop a strong foundation in the fundamental principles of art, such as color theory, composition, and perspective. Enroll in art classes, workshops, or online courses to deepen your understanding of these concepts. This knowledge will not only enhance your technical skills but also help you make more informed and intentional choices in your painting process.In addition to honing your technical abilities, it is crucial to nurture your creative vision and personal style. Experiment with different subject matter, themes, and approaches to find what inspires and resonates with you. Explore your own unique perspective and voice as an artist, and let it shine through in your work. This may involve experimenting with different styles, techniques, and subject matter until you discover what truly speaks to you.Another important aspect of becoming a painter is to build a strong artistic network. Connect with other artists, art enthusiasts, and industry professionals through local art communities, online forums, and social media platforms. Attend art events, workshops, and exhibitions to network, collaborate, and learn from others in the field. These connections can provide valuable insights, support, and opportunities for growth.As you progress in your artistic journey, it is essential to continuously challenge yourself and seek out new sources of inspiration. Engage in ongoing self-reflection and critical analysis of your work. Seek feedback from peers, mentors, and art professionals to identify areas for improvement and to refine your skills. Attend artist residencies, workshops, or artist-in-residence programs to immerse yourself in new environments and gain fresh perspectives.Furthermore, it is important to develop a strong portfolio that showcases your best work. This portfolio will be a crucial tool in securing commissions, gallery representation, and exhibition opportunities. Carefully curate your portfolio to highlight your technical skills, artistic vision, and unique style. Continuously update and refine your portfolio as you create new and compelling works.In the pursuit of becoming a painter, it is also essential to be financially savvy. Understand the business aspects of the art world, such as pricing your artwork, marketing your work, and navigating the art market. Explore various revenue streams, such as commissions, art sales, and teaching opportunities, to sustain your artistic practice.Finally, it is crucial to maintain a steadfast dedication and resilience in the face of challenges and setbacks. The path to becoming asuccessful painter is not without its obstacles. Be prepared to face rejection, criticism, and periods of self-doubt. Develop a strong support system, cultivate a growth mindset, and remain persistent in your pursuit of artistic excellence.In conclusion, becoming a painter requires a multifaceted approach that encompasses technical skill development, creative exploration, networking, continuous learning, and a deep-rooted passion for the art form. By embracing these key steps, you can embark on a fulfilling and rewarding journey towards realizing your dream of becoming a painter.。
关于计算机的常用词汇据360教育集团介绍:计算机基础知识:computer 电脑/电子计算机manipulate 操纵,操作information 消息/知识hand-hold 使携/手拿的calculator 计算器system 系统/体系scientific 科学的,系统的electronic 电子的machinery 机器/机关equipment 装备,设备dull 单调的,呆滞的network 网络circuit 电路/一圈/巡回switch 开关/电闸level 水平/标准status 状态binary 二进位的store 储存,储藏process 程序/过程character 字符sound 声音image 影像/图像programme 程序,计划logic inference 逻辑推理aid 帮助/援助instruction 指令convert 转变originality 创造力operate 操作,运转ENIAC 电子数值积分计算机vacuum 真空resistor 电阻器capacitor 电容器interference 干预technology 技术internal 内部的symbolic 代号language 语言span 跨越reliable 可靠的efficient 有效率的magnetic 一有磁性的Auxiliary 附加的/辅助物media 媒体headecimal 十六进制punched card tape 磁带memory 记忆/存储silicon 硅/硅元素chip 芯片terminal 终端机/终点/总站device 设备innovation 改革/创新external 外部的feature 特征component 元件/组件combination 联合/合并microprocessor 微处理器packed 包装的package 包裹/套装软件digital 数字的analog 模拟的hybrid 混合的discrete 离散的Vital 重要的/关键的monitor 显示器overwhelm 制服application 应用wire 电线,电报model 模型Versatility 多种变化,变通lump 使成块hardware 硬件stream 流resource 资源desktop 桌面cabinet 文件柜auxiliary storge 辅助存储器supercomputer 超级计算机minicomputer 小型计算机I/0 device 输入/输出设备system unit 系统部件cell 单元floppy disk 软盘consecutively 连续的/连贯的fix disk 硬盘CPU 中央处理器transmission 传送/传输操作系统与DOS操作基础storage space 存储空间Timer 计时器subdirectory 子目录Available 可用的structure 结构characteristic 特征/特性hierarchical 分层的Sophistication 复杂性issue 发行/放出Standard 标准backslash 反斜杠Online 联机the root directory 根目录Job Management 作业管理perform 执行Sequence 次序conjunction 联合Assess 评估procedure 过程Resource Management 资源管理tree 目录树Oversee vt.监督term 术语Control of I/0 Operation I/0 操作控制startup 启动Allocation 分配TSRs 内存驻留程序Undergo 经历/经受locate 定位Error Recovery 错误恢复sector 扇区Memory Management 存储器管理partition 分区interface 界面booting 自举streamlined 流线型的cluster 簇unleash 释放CMOS 互补金属氧化物体unhamperer 解脱emergency disk 应急磁盘spreadsheet 电子表格partition table 分区表Accessory 附件FAT 文件分配表Notepad 记事薄GUI 图形用户接口Macro Recorder 宏记录器command line 命令行Write 书写器icon 图标Paint-brush 画笔manual 手册modem 调制解调器dialog boxes 对话框Solitaire 接龙mechanism 机构/机械/结丰Reverse 挖地雷clipboard 剪贴板module 模块DDE 动态数据交换acronym 缩写字clumsy 笨拙的version 版本hot linked 映射的update 洲一级/更新real-mode 实模式internal command 内部命令standard mode 标准模式external command 外部命令directory 目录Pentium 俗称586 奔腾芯片sign-on 提示framework 框架/结构extension name 扩展名precedence 优先document 文档uppercase letter 大写字母workspace 工作lowercase letter 小写字母File Manager 文件管理volume label 卷标menu 菜单prompt 提示符Program Manager 程序管理器default 缺省值/默认值folder 卷宗symbol 符号divider 分配者cursor 光标subdivide 子分配者built-in 内置的tutorial 教程应用软件指南:maintenance 维护/维修Quit Batch 退出批处理install 安装.安置adapter 适配器advanced 高等的/在前的MDA 单显适配器copyright 版权/著作权CGA 彩色图形适配器duplication 副本,复制EGA 增强型图形适配器key letter 关键字VGA 视频图形阵列delete 删除destructive 破坏的/毁灭性的character string 字符串insert 寸击入/镶补verify 查证/证实bland 温和的/乏味的readable 可读的capacity 容量/能力attribute 属性/标志seek 搜寻/试图list 目录/名单/明细serial port 串行口sort 排序/分类/挑选loopback 回送alternate 交互的/轮流的specify 叙述/指定format 格式plug 插日argument 争论/引数/要旨ommunicate 沟通/传达match 使相配/使比赛peripheral 周边的/外设的path 路径/小路/轨道aspect 外观/方面pathname 路径名transfer 迁移/转移/传递head 头cache program 高速缓存程序relocation 再布置/变换布置subsystem 子系统/次要系统add 增加overall 全部的/全体的prune/graft 修剪/移植throughput 生产量/处延艳力resident 常驻程序numeric coprocessor 数学处理器compression 缩/缩小identify 识别/认明/鉴定reduce 减少/分解bargraph 长条图/直方图comment 批评/注解report 报告/报道extract 摘录/析取virus 病毒query 查询anti-virus 反病毒integrity 完整immunize 使免疫/赋予免疫性convert 使改变infection 传染/影响self-extractor 自抽出器original 最初的/原始的batch 批/成批result 结果/成绩/答案filename 文件名consider 考虑/思考/认为freshen (使)显得新鲜extra 额外的事物check 支票/检查restart 重新启动join 连接/结合detect 发现/察觉verbose 冗长/累赘的define 定义/详细说明edit 编辑编校suspicious 可疑的/疑惧的backup file 备份文件activity 活动/动作switch 开关转换warn 警告/注意beep 嘟嘟响present 现在的/出席的setting 设置exclusive 独占的/唯一的set mode 设置模式configuration 配置assume 假定/承担virus protection 防病毒density 密度scan 扫描细查inch 英寸signature file 签一名文件compatible 兼容的/能共处的editor 编辑器exception 例外/除外microcomputer 微机support 支持/支撑/援助retrieve 恢复/检索executable 可执行的/可运行的innovation 改革/创新documentation 文件manipulate 操纵/利用hit 打击/冲撞hardcopy 硬拷贝parameter 参数/媒介变数spell-checking 拼写检查evaluate 评估/评价thesaurus 辞典/同义词occur 发生/想到/存在merge 使合并/使消失valid 有效的/正当的function key 功能键buffer 缓冲区/缓冲familiarize 使熟悉/使熟知destination disk 目标盘wrap 包装/限制/包裹source disk 源盘blink 闪亮/闪烁overwrite 改写block 阻塞/封锁test 检验restore 恢复由backup制作的盘performance 绩效/表现/演出the space bar 空格键interrupt 中断accessory 附件/同谋group 团体/团retain 保持/留住/保有floppy drive 软盘驱动器locking 锁定hard drive 硬盘驱动器monitor 显示器parallel ports 并行口appropriate 适当的arrow 箭/箭头记号button 按钮highlight 加亮区/精彩场面optimize 使完善/优化horizontal 水平线/水平面indicator 指示器程序设计:creep 爬/潜行writing program 编写程序standardize 使标准化coding the program 编程simplify 单一化/简单化programming 程序revision 校订/修正programmer 程序员occupy 占领/住进logic 逻辑/逻辑学BASIC 初学者通用符号指令代码machine code 机器代码teaching language 教学语言debug DOS命令/调试simplicity 单纯/简朴compactness 紧凑的/紧密的timesharing system 分时系统description 描述/说明interactive language 交互式语言break 中断manufacturer 制造业者structure chart 结构图dialect 方言/语调the program flow 程序流expense 费用/代价manager module 管理模块uniformity 同样/划一worder module 工作模块archaic 己废的/古老的mainmodule 主模块sufficient 充分的/足够的submodule 子模块data processing 数据处理modify 修正/修改business application 商业应用outline 轮廓/概要scientific application 科学应用compose 分解lexical 字典的/词汇的code 代码non-programmer 非编程人员node 改为密码notation 记号法/表示法/注释pseudocode 伪代码verbosity 唠叨/冗长commas 逗点逗号record 记录documentation 文档subrecord 子记录flowchart/flow 程表/流程data division 数据部visual 视觉的procedure division 过程部represent 表现/表示/代表comprise 包含/构成structured techniques 结构化技术operator 运算符/算子straightforward 笔直的/率直的commercial package 商业软件包subroutine 子程序generator 产生器/生产者driver module 驱动模块mathematician 专家line by line 逐行operator 作符translate 翻译/解释forerunner 先驱modular 摸块化ancestor 祖宗cumbersome 讨厌的/麻烦的teaching programming 编程教学lengthy 冗长的/漫长的alter 改变flaw 缺点裂纹devclop 发达separate 各别的recompile 编译assist 帮助cycle 循环technician 技师remove 移动/除去straight line 直线category 种类/类项rectangle 长方形/矩形P-code p代码virtrally 事实上symology 象征学象征的使用register 寄存器to summaries 总之/总而言之by convention 按照惯例cyptic 含义模糊的/隐藏的diamond-shaped 菱形的bracket 括号decision 判断obviate 除去/排除terminal 终端机/终端的keyword 关键字card reader 阅读器underline 下划线translator program 译程序Programming 程序设计dec/binary 二进制source language 源语shift 变化/转移/移位machine language 机器overflow 溢出machine instruction 机器指令arithmetic 算术/算法computer language 计算机语言composite symbol 复合型符号assembly language 汇编语assignment 赋值floating point number 浮点数proliferation 增服high-level language 高级语pointer 指针natural language 自然语言array 数组矩阵,source text 源文本subscript 下标intermediate language 中间语言type conversion 类型转换software development 软件开发address arithmetic 地址运算map 映射/计划denote 指示/表示maintenance cost 维护费用subprogram 子程序legibility 易读性/易识别separate compilation 分离式编泽amend 修正/改善alphabetic 照字母次序的consumer 消费者digit 数字位数enormous 巨大的/庞大的numeric expression 数值表达式reliability 可信赖性/可信度tap 轻打/轻敲/选择safety 安全/安全设备print zone 打印区property 财产/所有权column 列correctness 正确functionality 机能semicolon 分号portable 叮携带的/可搬运的survey 概观altoggle 肘节开关task 作/任务declaration 宣告/说明source program 源程序mufti-dimension array 多维数组object program 目标程序数据库:transaction 交易/办理/执行query 查询license 执照/许可证/特许subschemas 子模式criminal 犯了罪的/有罪的individual 个体/个人conviction 定罪/信服/坚信employee 职员/受雇人员bureaus 局/办公处integrity 完整/正直insurance 保险/保险业/保险费duplicate 复制的/二重的retrieval 取回/恢复/修补interactive 交谈式security 安全/安全性audit 查帐/稽核integrity 完整/正直/廉正trail 痕迹/踪迹consume 消耗multiuse 多用户manually 用手full-fledged 喂养tedious 沉闷的/冗长乏味的compound document 复合文件DBMS 数据库管理系统recognizant 认识的/意识的consensus 一致/交感user manual 用户手册semantics 语义学bug 缺陷/错误impediment 妨碍/阻碍/阻止encrypt 加密/译成密码intuitively 直觉的malicious 环恶意的/恶毒的module 模块/组件bottleneck 瓶颈schema 轮廓/概要/图解mainstream 主流proposal 建议spatial 空间的/空间性的tailor 定制/制作/缝制relevant 有关联的/中肯的plausible 似真实的/似合理的urgency 紧急/催促virtually 事实上optimization 最佳化impracticably 不能实validation 确认flaw 缺点/裂纹/瑕疵typically 典型的/象征性的assumption 假定/视为当然之事index 索引/做索引duration 持续时间/为期component 组件/成分intolerably 难耐的程度temporal 当时的/现世的abort 流产/失败semantics 语义学rigorous 严厉的/严酷的/苛刻的interval 时间间隔criterion 标准/准据/轨范catalogue 目录/编入目录consistency 一致性/坚固性/浓度cabinet 橱柜/内阁adopt 采用/收养illustration 例证/插图serialization 连载长篇efficient 有效率的/能干的log 日志/记录clerical 事务上的/抄写员的focus 焦点/焦距access 进入twin 双胞胎中人warehouse 大商店/仓库protocol 协议wholesale 批发conflict 神突/矛盾chore 零工/家务negotiate 商议/谈判/谈妥mode 模式/模态drag 拖拉/拖累long-duration 长期architects 建筑师short-duration 短期partition 分割/隔离物ascend 上升/追溯/登高.inherent 固有的/与生俱来的descend 下降/传下necessitate 迫使/使成为必需dimensional 空间的versa 反physical organization 物理组织operator 操作员数字电路:digital circuit 数字电路inclusive 一包含的/包括的logic 逻辑bit 少量gate 逻辑门multibit 多位logical methodology 逻辑方法arithmetic operation 算术运算Boolean algebra 布尔代数bus 总线two-state 两态data bus 数据总线logical multiplication 逻辑乘simultaneously 同时地logical addition 逻辑加parallel register 并行寄存器logical complementation 逻辑非serial register 串行寄存器logical function 逻辑函数shift register 移位寄存器inverter 反相器transistor 晶体管electromechanical calculator 电动式计算器diode 二极管logic symbol 逻辑符号resistor 电阻器electromagnet 电磁铁logic circuit 逻辑电路energize 使活跃/激励Flip-flop 发器armature 电枢counter 计数器relay 电器adder 加法器mechanical latch 机械式,logic variable 逻辑变量set 置位logic operation 逻辑运算reset 复位characteristic 特征/特性figure 图the SET output 置位输出端conjunction(logical product) 合取the RESET input 复位输入端disjunction(logical sum) 析取first-level 一级active 有效的negation(NOT) 反(非)inactive 无效的AND gate 与门construct 构造/设想truth table 真值表resident program 常驻程序power 功率/乘幂utility 公用程序/实用condition 条件diskcopy 磁盘拷贝命令verbalize 以语言表现/唠叨exception 例外vice Vera 反之亦然batch 批/成批the AND function “与”函数specify 指定/说明the OR function “或”函数discrepancy 相差/差异/差别the NOT function “非”函数trigger 触发器exemplify 例证/例示representative 代表/典型硬件基础:microelectronics 微电子学adaptively 适合的/适应的actuator 主动器compensate 偿还/补偿integrated 集成的parasitic 寄生的arithmetic 算术/算法wobble 摆动/不稳定crossroads 交又路focal 焦点的/在焦点上的ROM 只读存储器eliminate 排除/除去RAM 随机存取存储器cornstalk 串音permanently 永久的/不变的affinity 密切关系/强烈的吸引Volatile 可变的/不稳定的stem 柄/堵塞物notepad 记事本introspection 内省/反省microprocessor 微处理器mechanism 机械/机理gateway 门/通路portability 一携带/轻便coprocessor 协处理器configuration 配置floating-point 浮点flexibility 适应性/弹性upgrade 使升级algorithms 运算法则optional 选择的/随意的channel 通道/频道bi-directional 双向性keystroke 键击simultaneous 同时发生的typematic 重复击键的cache 高速缓冲存储器comprise 包含/构成percentage 百分比/部分precommendation 预补偿controller 控制器track 磁轨intercept 截取/妨碍boot 启动significantly 重要地/有效地benchmark 基准/评效migration 移往/移动merit 优点/价值compact 紧凑的/紧密的restriction 限制/限定/约束digitally 数位intrinsic 本质的/原有的dip 双排直插封装Boolean 布尔逻辑/布尔值distortion 扭曲/变形imperative 命令式的playback 重现/录音再生nontrivial 不平常的robustness 健康的/强健的circumvent 绕行/陷害reliability 可靠性/可信赖性decentralize 使分散/排除集中resolvability 可移动性intelligent 智能的/聪明的counterpart 副本/配对物automatically 自动地/机械地archival 关于档案的innovation 改革/创新magneto 磁发电机synonym 同义字cylinder 柱面prototype 原型photodetector 光感测器paradigm 范例/模范predefined 预先确定microchip 微处理器split 分散的core 争论的核心tradeoff 交换,协定extended memory 扩充内存bootdevice 引导设备picture processing 图像处理reside 住/居留/属于sensor 传感器optical disk 光盘WS1 晶片规模集成laser 激光VLSI 超大规模集成storage densities 存储密度hiss 嘶嘶声modulate 调整/调制unveil 揭开/揭幕multiassociative processing 多关联处理技术workload 工作负荷计算机网络与分布式系统:network 网络zap 意志/活力coordinate 同等的/(使)协调hassle 争论minicomputer 小型计算机legacy 传统的facility 设备/容易Macintosh 大苹果机LAN 局部区域网络workstation 工作站irrespective 不顾的/无关的catapulting 发射机弹弓distributed network 分布式网络meteorological 气象学的central machine 中央主机centralization 集中appropriate 适当的immune 免疫的/免除的software packages 软件包immunity 免疫/免疫性meaningful 意味深长的equatorial 近赤道的,赤道的ring network 封闭网络discipline 训练/惩罚stress 重点/紧迫homogeneity 同种/同质open system 开放系统improvisation 即兴而作/即席演奏backup 做备份ultimately 终极/根本interconnection 互联historically 历史的/史实的quotation 引用语payroll 工资单catalog 目录/型录browser 浏览器bulletin 公告,neutral 中立者/中立国approach 接近/动手处理enhance 提高/加强impractical 不实际的endorse 支持/赞同crucial 决定性的/重要的accelerate 加速operability 相互操作性mission 任务/使命scaleable 可攀登的/可剥掉的critical 批评的/决定性的tightly 紧紧地/坚固地inventory 存货清单longevity 长命/长寿/寿命administrative 行政的/管理的evaluating 评估strategy 策略dispersed 被分散的remote 远程incremental 增加的monitoring 监听intervention 插入/介入conventional program 常规程序host 主机/主人supervisory 管理的/监督的warrant 凭证/正当理由versatile 万用的peripherals(计算机)辅助设备collaborate 合作realm 王国/领域download 下载analogize 以类推来说明proliferate 增殖/激增quadrate 求积/矩/弦website web地址amplitude 广阔/充足/增幅OSI 开放系统互联network management 网络管理product development 产品开发signal level 信号电平integrated network 集成网络object-oriented 面向对象file server 文件服务器object definition 对象定义mouse 鼠标fault isolation 故障隔离click 单击entry 登录/入口database system 数据库系统DTE 数据终端设备centralized system 集中式系统paralleled-to-serial 并串decentralized system 分散式系统serial-to-paralleled 串并distributed system 分布式系统Universal Synchronous 通用同步workstation 工作站Asynchronous Receiver 异步接收coordinate 坐标/同等的transmitter 发送器multipoint data 多点数据data stream 数据流FEP 前端处理机modulator 数传机arithmetic logic unit 算术逻辑部件keyboard 键盘printer 打印机skitter 磁盘statistical 统计的joystick 游戏棒/操纵杆software 软件category 种类simulate 模拟,模仿handle 控制interpret 解释feedback 反馈instrument 工具manufacture 制造CAD 计算机辅助设计engineer 工程师draft 草稿graphics 图形video 影像robotic 机器人学automation 自动化word processing 字处理text 文本communication 通讯electronic-mail 电子邮件teleconferencing 电话会议telccommunicating 远程通讯database 数据库CAI 计算机辅助教学transistor 晶体管DOS 磁盘操作系统RAM 随机存取存储器mouse 鼠标intense 强烈/紧张floppy 松软的fix 牢固的write-protect 写保护drive 驱动器mechanics 机械学access 访问byte 比特mega 兆decimal 十进制octal 八进制storage 存储器weight 权code 代码ASCII 美国信息交换标准代extended 扩充的/长期的voltage 伏特integer 整数negative 负的absence 缺席convenience 便利waveform 波形zone 区vendor 厂商/自动售货机implement 工具/器具quantity 数量rigid 硬的fragile 易脆的susceptible 易受影响的medium 媒体shutter 快门general-purpose 通用theory proving 定理证明information retrieval 信息检索persona computer 个人计算机time-consuming 费时的routine task 日常工作logical decision 逻辑判断programmable 可编程的rewire 重新接线generation 代unreliable 不可靠的。
我想发明智能魔方英语作文I've always had this idea of inventing a smart Rubik's Cube. You know, the kind that can solve itself or give you hints on how to solve it. It would be so cool to have a cube that can adapt to your skill level and challenge you accordingly. I can imagine how much fun it would be to play with a cube like that.Imagine being able to connect your smart Rubik's Cube to your phone and track your progress. You could compete with friends, join online competitions, and maybe even learn new solving techniques from the app. It would be like having a personal Rubik's Cube coach right in your pocket.I think it would be really interesting to see how a smart Rubik's Cube could be used in educational settings.It could be a great tool for teaching problem-solving skills, spatial reasoning, and even algorithms. Kids could learn through play and develop valuable skills without even realizing it.One of the coolest things about a smart Rubik's Cube would be the potential for customization. You could change the colors, patterns, and even the size of the cube to make it uniquely yours. It would be a great way to express your personality and stand out in the crowd of cubers.I can't help but wonder how a smart Rubik's Cube would impact the speedcubing community. It could revolutionize the way competitions are held and push cubers to new levels of skill and strategy. I bet it would be a game-changer for the sport.。
TECHNICAL ADVANCEVisualization of protein interactions in living plant cells using bimolecularfluorescence complementationMichael Walter1,Christina Chaban2,Katia Schu¨tze2,Oliver Batistic1,Katrin Weckermann3,Christian Na¨ke2,Dragica Blazevic1, Christopher Grefen2,Karin Schumacher3,Claudia Oecking3,Klaus Harter2,*and Jo¨rg Kudla1,*1Institut fu¨r Botanik und Botanischer Garten,Molekulare Entwicklungsbiologie der Pflanzen,Universita¨t Mu¨nster, Schlossplatz4,48149Mu¨nster,Germany,2Botanisches Institut,Universita¨t zu Ko¨ln,Gyrhofstr.15,50931Ko¨ln,Germany,and3ZMBP,Pflanzenphysiologie,Universita¨t Tu¨bingen,Auf der Morgenstelle1,72076Tu¨bingen,GermanyReceived24June2004;revised6August2004;accepted12August2004.*For correspondence(faxþ492518323311;e-mail jkudla@uni-muenster.de:faxþ492214707765;e-mail klaus.harter@uni-koeln.de).SummaryDynamic networks of protein–protein interactions regulate numerous cellular processes and determine theability to respond appropriately to environmental stimuli.However,the investigation of protein complexformation in living plant cells by methods such asfluorescence resonance energy transfer has remainedexperimentally difficult,time consuming and requires sophisticated technical equipment.Here,we report theimplementation of a bimolecularfluorescence complementation(BiFC)technique for visualization of protein–protein interactions in plant cells.This approach relies on the formation of afluorescent complex by two non-fluorescent fragments of the yellowfluorescent protein brought together by association of interacting proteinsfused to these fragments(Hu et al.,2002).To enable BiFC analyses in plant cells,we generated differentcomplementary sets of expression vectors,which enable protein interaction studies in transiently or stablytransformed cells.These vectors were used to investigate and visualize homodimerization of the basic leucinezipper(bZIP)transcription factor bZIP63and the zincfinger protein lesion simulating disease1(LSD1)fromArabidopsis as well as the dimer formation of the tobacco14-3-3protein T14-3c.The interaction analyses ofthese model proteins established the feasibility of BiFC analyses for efficient visualization of structurallydistinct proteins in different cellular compartments.Our investigations revealed a remarkable signalfluorescence intensity of interacting protein complexes as well as a high reproducibility and technicalsimplicity of the method in different plant systems.Consequently,the BiFC approach should significantlyfacilitate the visualization of the subcellular sites of protein interactions under conditions that closely reflectthe normal physiological environment.Keywords:bimolecularfluorescence complementation,protein–protein interaction,intracellular localization,bZIP transcription factor,14-3-3proteins,LSD1.IntroductionThe regulation and execution of biological processes requires specific interactions of numerous proteins.Tightly regulated protein interaction networks mediate cellular responses to environmental cues and direct the implemen-tation of developmental programs.The selectivity of protein–protein interactions and their appropriate temporal and spatial regulation determine the developmental potential of the cell and its response to endogenous and exogenous sig-nals.On the molecular level differential protein–protein interactions are thought to determine the operation of com-plex regulatory circuits and signal transduction systems. The complete sequencing of an increasing number of eukaryotic genomes has provided a wealth of information about the number and complexity of protein functions428ª2004Blackwell Publishing Ltd The Plant Journal(2004)40,428–438doi:10.1111/j.1365-313X.2004.02219.xrequired to build up an organism.However,the regulation and interplays of these proteins remain to become explored in order to appreciate the molecular mechanisms of their action.Several methods have been developed to identify, examine and visualize protein interactions and protein com-plexes in living cells.Among them,the yeast two-hybrid system has significantly advanced the speed and extent of protein interaction studies.However,this system bears intrinsic limitations as for example systematic‘false-positive’and‘false-negative’interactions and,moreover,usually combines protein pairs in a heterologous environment(Field and Song,1989;Stephens and Banting,2000).The most widely used approach for the visualization of protein inter-actions in living cells isfluorescence resonance energy transfer(FRET)between spectral variants of the green fluorescence protein(GFP)fused to the associating proteins (Chen et al.,2003;Periasamy,2000).However,to enable observation and quantification of small alterations influor-escence emission,the GFPfluorophores have to join in close spatial proximity and the fusion proteins generally have to be expressed in high levels.Furthermore,verification,whether changes influorescence emission are caused by energy transfer,requires complicated irreversible photobleaching or fluorescence lifetime imaging techniques(Chen et al.,2003; Periasamy,2000).However,the instrumental equipment necessary for these techniques is not widely available and FRET requires intensive methodical training.For these rea-sons reports about FRET-based protein–protein interaction investigations in living cells have remained rare especially in plant science(Immink et al.,2002;Ma´s et al.,2000;Shah et al.,2002;Vermeer et al.,2004).Alternatively,protein interactions can also be investigated in vivo if the protein complex formation can be visualized by the restoration of a detectable activity.In this regard,the principle of intragenic complementation of the lacZ locus from Escherichia coli was adapted to detect protein interac-tions(Rossi et al.,1997;Ullmann et al.,1967).In this experimental system the detection of protein–protein inter-actions by restoration of b-galactosidase activity was enabled by using b-galactosidase fragments,which could associate only when fused to interacting proteins.Similarly, fragments of the dihydrofolate reductase have been used in protein interaction studies based on complementation of protein function(Pelletier et al.,1998).However,these techniques require the application of extrinsicfluorophores to visualize the complex formation.An alternative experimental approach for the visualization of protein interactions is based on the formation of a fluorescent complex by fragments of the enhanced yellow fluorescent protein(YFP)when brought together by the interaction of two associating partners fused to these frag-ments.Recently,Kerppola and colleagues(2002)reported a proof-of-concept for such an approach for the investigation of protein interactions in living mammalian cells and designa-ted this technique as bimolecularfluorescence complemen-tation(BiFC).The unique characteristic of the BiFC approach is that the bright intrinsicfluorescence of the bimolecular complex allows direct visualization of the complex formation in living mammalian cells.Moreover,by analyzing the interactions between members of the basic leucine zipper (bZIP)and Rel transcription factor families,the BiFC approach provided direct evidence of the intracellular locations where the protein association occurs(Hu et al.,2002).The applica-tion of the BiFC approach has recently been extended to the investigation of the interaction pattern and intracellular localization of G-protein complexes in mammalian cells and Dictyostelium discoideum(Hynes et al.,2004)and to the visualization of1-aminocyclopropane-1-carboxylase syn-thase heterodimer formation in E.coli(Tsuchisaka and Theologis,2004).Furthermore,by introducing a large number of different GFP variants the technique was extended to multicolor BiFC,which allows the direct visualization of multiple protein interactions within the same cell(Grinberg et al.,2004;Hu and Kerppola,2003).In this report we describe the generation of several sets of plant-compatible BiFC vectors.We used these vectors for investigating the interaction of plant nuclear and cytoplas-mic proteins in different plant systems.Our study attests the general applicability of the BiFC technique and that this assay represents an efficient and convenient tool to investi-gate protein–protein interactions in living plant cells.ResultsGeneration of plant-compatible BiFC transformation vectorsTo develop the BiFC technology for the visualization of protein–protein interactions in living plant cells,we con-structed four pairs of vectors(Figure1;for further details see Experimental procedures and Supplementary material). These vectors have been designated pSPYNE and pSPYCE (for split YFP N-terminal/C-terminal fragment expression) respectively.Each vector pair enables the expression of proteins of interest fused either to the N-terminal155amino acids(YFP N)or to the C-terminal86amino acids of YFP (YFP C;Hu et al.,2002).Moreover the plasmids contain either a c-myc(pSPYNE)or HA(pSPYCE)affinity tag for detection of fusion protein expression in cell extracts(Figure1).The binary pSPYNE-KAN and pSPYCE-BAR vectors enable the expression of YFP fragment-fused genomic DNA or of YFP-fragment constructs driven by any promoter of interest (Figure1).Strong and constitutive expression of fusion proteins in plant cells is ensured by the binary pSPYNE-35S and pSPYCE-35S plasmids which contain the35S promoter of the cauliflower mosaic virus.For selection of transgenic plants pSPYNE-KAN and pSPYNE-35S carry a nos promoter-driven kanamycin resistance gene(nptII),whereas pSPYCE-BAR and pSPYCE-35S harbor the bar gene conferringProtein interaction in plant cells429ªBlackwell Publishing Ltd,The Plant Journal,(2004),40,428–438insensitivity to the herbicide glufosinate (Figure 1).In addi-tion,we generated two additional sets of vectors based on pUC19that are specially designed for transient plant cell transformation approaches.pUC-SPYNE and pUC-SPYCE contain the entire expression cassette of pSPYNE-35S or pSPYCE-35S ,respectively,and harbor a more variable multi-cloning site (MCS)(Figure 1).Furthermore,in a second set the entire MCSs of pUC-SPYNE and pUC-SPYCE were re-placed by the Gateway conversion cassette providing the attR1and attR2recombination sites for use with the Gate-way cloning system (pUC-SPYNE G ,pUC-SPYCE G ,Figure 1).BiFC analysis of Arabidopsis nuclear bZIP63To address the feasibility of BiFC for visualization of protein–protein interaction in living plant cells we first choose a member of the Arabidopsis bZIP factor family (bZIP63,AGI:At5g28770)as a model protein.bZIP63belongs to subfamily C of Arabidopsis bZIP factors (Jakoby et al.,2002).This transcription factor binds to promoter elements containing the CACGTG or GACGTC sequence in vitro and is localized to the nucleus of plant cells (Na ¨ke,2001).Moreover,bZIP transcription factors are known to form homodimers and heterodimers via the C-terminal leucine zipper domain (Siberil et al.,2001).To first investigate the interaction potential of bZIP63by an independent experimental approach,we performed an interaction analysis in the yeast two-hybrid system.As shown in Figure 2,bZIP63formed homodimers in vivo as demonstrated by the growth of transformants on interaction selective medium and induction of b -galactosidase reporter activity above background level.To corroborate thattheFigure 1.Schematic representation of plant-compatible BiFC vectors.(a)pSPYNE-Kan and pSPYCE-Bar.(b)pSPYNE-35S/pUC-SPYNE and pSPYCE-35S/pUC-SPYCE.(c)pUC-SPYNE G and pUC-SPYCE G .Details of the plasmid construction and vector back bones are given in Experimental procedures and in Figures S1and S2.c-myc,c-myc affinity tag;HA,hemagglutinin affinity tag;MCS,multi-cloning site;35S,35S promoter of the cauliflower mosaic virus;NosT,terminator of the Nos gene;YFP N ,N-terminal fragment of YFP reaching from amino acid (aa)1to 155;YFP C ,C-terminal fragment of YFP reaching from amino acid 156to 239;attR1-Cm R -ccdB-attR2,Gateway conversion cassette.430Michael Walter et al.ªBlackwell Publishing Ltd,The Plant Journal ,(2004),40,428–438homodimer formation is mediated by the leucine zipper we introduced two point mutations in the bZIP63sequence which changed Leu188and Leu195into Pro.Leu188and Leu195are the first two hydrophobic amino acids in the C-terminal zipper-forming amphipathic a -helix of the bZIP63monomers.Therefore,conversion of these positions into prolines is likely to interfere with the dimerization potential of the protein (Siberil et al.,2001).The combination of wild type bZIP63with the mutated version (bZIP63PP )reduced expression of the reporter genes (Figure 2).It is noteworthy that in yeast mutation of Leu188and Leu195to proline does not completely abolish homodimerization of bZIP63(Fig-ure 2).In summary,these data indicate that the leucine zipper domain is predominantly responsible for bZIP homodimer formation in yeast.We next attempted the direct visualization of homodime-rization in living plant cells.To this end,we transiently transfected Arabidopsis cell culture protoplasts with various pUC-SPYNE /pUC-SPYCE constructs of bZIP63and,in addi-tion to microscopic analysis,quantified the fluorescence intensity.Whereas cells transfected with single plasmids and any combination with empty vectors produced no or only background fluorescence,a strong signal was observed when bZIP63-YFP N was co-expressed with bZIP63-YFP C (Figure 3a,b).Significantly weaker fluorescence signals were observed when combinations of bZIP63PP with wildtype bZIP63were transfected,thereby reflecting reduction in homodimerization by these mutations (Figure 3a,b).Gener-ally the number of BiFC signal-emitting protoplasts was about the half compared with cells expressing full-length GFP fusion proteins.This difference corresponds to the reduced efficiency when more than one construct is used for transfection.To test the functionality of the binary BiFC vectors in planta ,the wild type bZIP63and bZIP63PP cDNAs were cloned into pSPYNE-35S and pSPYCE-35S,respectively.The constructs were delivered into leaf cells of tobacco (Nicoti-ana benthamiana )by Agrobacterium infiltration (Voinnet et al.,2003;Witte et al.,2004).Similar to the situation in Arabidopsis protoplasts strong YFP fluorescence was observed when wild type combinations of bZIP63were expressed (Figure 4a,panels I,II).Pairwise expression of bZIP63and bZIP63PP ,bZIP63PP alone or in combination with the YFP fragments induced no or only weak fluorescence signals (Figure 4a,panels I,II and data not shown).Using HA-and c-myc-tag-specific antibodies the expression of all fusion proteins in tobacco cells was demonstrated (Fig-ure 4a,panel III).Notably,we observed that the transforma-tion efficiency of Agrobacterium -infiltrated tobacco cells strongly depends on the constructs used.For instance,whereas 80%of epidermal cells which have been infiltrated with bZIP63-YFP N and bZIP63-YFP C -carryingAgrobacteriaFigure 2.Homodimerization of bZIP63and bZIP63PP in yeast.The indicated Gal4DNA-binding domain (BD)and activation domain (AD)constructs were transformed into yeast strain PJ69-4A.Transformants were assayed for the activity of protein–protein interaction reporting genes either by growth in decreasing densities (narrowing triangle)on selective medium (I:CSM-L,W,A)or determination of b -galactosidase activity (II).CSM-L,W (I)depicts a dilution series on non-selective control plates.bZIP63PP represents a mutated version of bZIP63in which Leu188and Leu195have been mutated to Pro.Protein interaction in plant cells 431ªBlackwell Publishing Ltd,The Plant Journal ,(2004),40,428–438showed BiFC-induced fluorescence,LSD1homodimer formation (see Figure 7)was observed in only 20%of the cells.In transfected Arabidopsis protoplasts and infiltrated tobacco leaves the homodimerization-induced YFP fluores-cence appeared exclusively inside the nucleus which is in agreement with the observation that bZIP63-GFP and bZIP63PP -GFP are nuclear proteins (Figures 3c and 4b).BiFC analysis of proteins in the cytoplasm of plant cells To further extend the applicability of BiFC beyond bZIP transcription factors we analyzed a 14-3-3protein (isoform T14-3c from N.tabacum ;GenBank:NTU91724)and the zinc finger protein LSD1from Arabidopsis thaliana (Dietrich et al.,1997)by BiFC analyses in Arabidopsis protoplasts and Agrobacterium -infiltrated tobacco leaves respectively.14-3-3proteins form a conserved family of eukaryotic polypeptides that were the first signaling molecules identi-fied as discrete phosphoserine/phosphothreonine-binding modules.They associate to homodimers or heterodimers with a saddle-shaped structure,with each monomer forming an extended groove that allows binding of the phosphoryl-ated sequence motif (Rittinger et al.,1999;Wu ¨rtele et al.,2003).Dimerization of 14-3-3proteins occurs via their N-terminal region.Accordingly,yeast two-hybrid analyses revealed that an N-terminally truncated version of T14-3cisFigure 3.BiFC visualization of bZIP63dimeriza-tion in transiently transfected Arabidopsis thali-ana cell culture protoplasts.(a)Epifluorescence (I)and bright field images (II)images of Arabidopsis cell culture protoplasts co-transfected with constructs encoding the indi-cated fusion proteins.(b)Quantification of fluorescence intensities in transiently transfected Arabidopsis cell culture protoplasts.Fluorescence intensity (arbitrary units)was determined using the Metamorph software.The mean and standard deviation of three independent measurements are shown.(c)bZIP63-GFP and bZIP63PP -GFP are localized to the nucleus.Epifluorescence (I)and bright field (II)images of protoplasts transfected with con-structs expressing the indicated fusion proteins.Scale bars,20l m.432Michael Walter et al.ªBlackwell Publishing Ltd,The Plant Journal ,(2004),40,428–438no longer able to homodimerize (Jaspert and Oecking,2002).For BiFC studies the cDNAs encoding wild type T14-3c (T14)and the mutant version (T14D N)were cloned into pUC-SPYNE and pUC-SPYCE or pSPYNE-35S and pSPYCE-35S ,respectively,and transformed into either Arabidopsis pro-toplasts or tobacco leaf cells.Upon co-expression of T14-YFP N and T14-YFP C ,strong YFP fluorescence was detected throughout the cytoplasm and the nucleus in both systems indicating that homodimerization occurs in both compart-ments (Figures 5a and 6a).In transformed epidermal cells of tobacco the cytoplasmic fluorescence typically appears as a thin area between the cell wall and the turgescent vacuole.Plasmolysis of the cells by treatment with 500m M mannitol demonstrated the localization of the BiFC signal in the cytoplasm (data not shown).Importantly,the localization of homodimer formation corresponds to the subcellular distri-bution of T14-3c,when expressed as GFP fusion in proto-plasts or tobacco leaf cells under the control of the 35S promoter (Figures 5b and 6b).In contrast,expression of neither the N-terminal truncated form T14D N fused to the YFP fragments nor any combination of wild type with the truncated version resulted in a YFP signal (Figures 5a and 6a).Western blot analysis confirmed that the expressionofFigure 4.BiFC visualization of bZIP63dimerization in Agrobacterium -infiltrated tobacco (Nicotiana benthamiana )leaves.(a)Epifluorescence (I)and bright field (II)images of epidermal leaf cells infiltrated with a mixture of Agrobacterium suspensions harboring constructs encoding the indicated fusion proteins.In addition,the second panel from above shows a confocal image of BiFC-induced bZIP63dimerization.For technical details of infiltration see Experimental procedures.The expression of the proteins (III)is demonstrated by immunodetection with anti-HA (a -HA)antibodies for YFP C fusions and anti-c-myc (a -c-myc)for YFP N fusions.*,degradation product.(b)bZIP63-GFP and bZIP63PP -GFP are both localized to the nucleus of plant cells.Epifluorescence images of Agrobacterium -infiltrated tobacco (N.bent-hamiana )epidermal cells are shown.Scale bars,50lm.Figure 5.The tobacco 14-3-3protein T14-3c interacts in Arabidopsis cell culture protoplasts.(a)Bright field (I)and epifluorescence (II)images of Arabidopsis cell culture protoplasts co-transfected with constructs encoding the indicated fusion proteins.(b)T14-3c-GFP is localized to the cytoplasm and nucleus.Bright field (I)and epifluorescence (II)images of protoplasts transfected with a construct expressing T14-3c-GFP are depicted.(c)Demonstration of protein expression by immunodetection with anti-HA (a -HA)antibodies for YFP C fusions and anti-c-myc (a -c-myc)for YFP N fusions.Extracts from protoplasts co-transfected with the constructs indicated in (a)are shown (lanes 1–4).Scale bars,20l m.Protein interaction in plant cells 433ªBlackwell Publishing Ltd,The Plant Journal ,(2004),40,428–438the non-interacting T14-3c fusion proteins was comparable with the expression of the interacting form of T14-3c (Figures 5c and 6c).LSD1is a small zinc finger protein that forms homodimers in vivo and functions as a negative regulator of programmed cell death in plants (Dietrich et al.,1997;Epple et al.,2003).As observed for the T14-3c homodimer,co-expression of LSD1-YFP N and LSD1-YFP C induced strong fluorescence in the cytoplasm and the nucleus of infiltrated tobacco cells,whereas control pairs gave no or only a very weak YFP signal (Figure 7a and data not shown).Again,the location of LSD1homodimer formation is identical to the localization of LSD1-YFP when expressed under the control of the 35S promoter in transgenic Arabidopsis plants (Figure 7b).To test the specificity of LSD1protein association,we analyzed the interaction of bZIP63and LSD1in co-transformed tobacco leaf cells.This protein pair was chosen because in the yeast two-hybrid system bZIP63does not interact with LSD1(Na ¨ke,2001).As shown in Figure 7,co-expression of bZIP63and LSD1in any combination revealed no fluores-cence,although the fusion proteins were expressed.DiscussionIn this report we establish BiFC as a very efficient technology for the analysis of protein–protein interactions in liv-ing plants cells.The vectors described are readily suitable for BiFC analyses in Arabidopsis protoplasts and Agrobacterium -infiltrated tobacco leaves.They also allow determination of protein expression levels by Western blot analysis.The binary BiFC vectors will also enable interaction and protein complex formation studies in transgenic plants.However,at high expression levels the free YFP frag-ments sometimes tend to associate non-specifically,therebyFigure 7.LSD1forms homodimers in planta but does not interact with bZIP63.(a)Epifluorescence (I)and bright field (II)images of epidermal leaf cells infiltrated with a mixture of Agrobacterium suspensions harboring constructs encoding the indicated fusion proteins.For technical details of infiltration see Experimental procedures.The expression of the proteins (III)is demonstrated by immunodetection with anti-HA (a -HA)antibodies for YFP C fusions and anti-c-myc (a -c-myc)for YFP N fusions.(b)LSD1-YFP is localized to the nucleus and cytoplasm of Arabidopsis cells.Epifluorescence (I)and bright field (II)images of hypocotyl cells from transgenic Arabidopsis plants which express LSD1-YFP are depicted.Scale bars,50lm.Figure 6.BiFC visualization of T14-3c dimerization in tobacco leaves.(a)Confocal (I)and bright field (II)images of epidermal leaf cells from Nicotiana benthamiana infiltrated with a mixture of Agrobacterium suspen-sions harboring constructs encoding the indicated fusion proteins.For technical details of infiltration see Experimental procedures.(b)T14-3c-GFP is localized to the cytoplasm and nucleus.Confocal (I)and bright field (II)images of stably transformed N.tabacum epidermal cells are depicted.(c)Demonstration of protein expression by immunodetection with anti-HA (a -HA)antibodies for YFP C fusions and anti-c-myc (a -c-myc)for YFP N fusions.Extracts from tissue co-transformed with constructs co-expressing either T14-YFP N and T14-YFP C (lane 1)or T14D N-YFP N and T14D N-YFP C (lane 2)are shown.434Michael Walter et al.ªBlackwell Publishing Ltd,The Plant Journal ,(2004),40,428–438generating backgroundfluorescence(Figure3a,b).This problem can be circumvented,when a non-interacting fusion protein is used in the control experiments as exemplarily shown for the T14D N and bZIP63/LSD1pairs,or if the expression level is reduced by the use of a less active promoter.Our BiFC analyses of interactions among bZIP63,T14-3c and LSD1in living plant cells illustrate several significant advantages of this technique(Hu et al.,2002).(i)Protein interactions using BiFC are visualized in the normal envi-ronment of the plant cell.Several restrictions inherent to interaction approaches in heterologous systems(e.g.yeast), as for instance missing plant-specific post-translational protein modifications or incorrect subcellular localization, are overcome by BiFC.From a technical point of view the detection of BiFC-generated interactions does not require accessibility of the protein complex to extrinsicfluorophores and does not require the instrumental equipment necessary for FRET including subsequent complex data processing.(ii) Compared with published FRET data(Immink et al.,2002; Ma´s et al.,2000;Shah et al.,2002;Vermeer et al.,2004)the BiFC signals observed in our study using the strong viral35S promoter for the expression of the fusion proteins appear to be very parison of the microscopic exposure times suggest that for a given protein interaction the BiFC fluorescence intensity can reach about30%of the signal intensity of the corresponding full-length GFP.From these data we conclude that BiFC is very sensitive and may allow the detection of interactions when the proteins are expressed at lower levels under the control of native promoters.As demonstrated by our study and recent reports (Grinberg et al.,2004;Hynes et al.,2004)BiFC-generated fluorescence signals can also be quantified.Thus,the ability of a protein to form different complexes with different interaction partners can be quantitatively determined at the cellular or even the subcellular level(Grinberg et al.,2004). (iii)The most appealing advantage of BiFC is that protein interaction occurs in the genuine compartment of the proteins examined.Arabidopsis bZIP63is a nuclear tran-scriptional regulator and homodimer formation is exclu-sively detected by BiFC inside the nucleus.When expressed under the control of the viral35S promoter,tobacco T14-3c and Arabidopsis LSD1accumulate in the cytoplasm and in the nucleus.Again,BiFC dimer formation strictly coincides with observed localization in these compartments.This should enable the identification of the subcellular distribu-tion of interaction between regulatory proteins like tran-scription factors and signaling components and may immediately provide information about the functional role of the association.For instance,in mammalian HEK-293cells BiFC analyses revealed that the different b subunits of G proteins direct the corresponding bc signaling complexes to alternative subcellular locations(Hynes et al.,2004).Further-more in,mammalian cell lines using BiFC Kerppola and colleagues demonstrated that the cytoplasmic transcription factor Mad4was recruited to the nucleus through dimeriza-tion with Max(Grinberg et al.,2004)and were able to identify the intramolecular region responsible for the differ-ential intracellular distribution of ATF2/Jun heterodimers (Hu et al.,2002).(iv)Formation of the BiFC protein complex occurs through a multistep pathway in vitro and very likely in vivo(Hu et al.,2002).The initial steps are mediated by contacts between the proteins fused to the YFP fragments and are reversible.At this initial state complex formation can be competed by alternative interaction partners.Afterwards the initial complex is stabilized by the association of YFP fragments and further exchange of protein components is inhibited(Hu et al.,2002).Therefore,the kinetics of BiFC formation allows the detection of weak and transient protein complexes in vivo but has the disadvantage that shifts in the equilibrium between putative alternative complexes may hardly be detectable after initial formation.However,we expect that the dynamics of BiFC protein complex formation is still maintained or can be still modified by the cellular protein folding and degradation machinery.(v)The structural background of protein complex formation can likewise be investigated by BiFC.For instance,truncation of the N-terminal region of the tobacco T14-3c protein or point mutations within the C-terminal leucine zipper of Arabidop-sis bZIP63identified these domains to be responsible for homodimerization in plant cells.(vi)In perspective,the BiFC approach enables the identification of signals that induce the formation of protein complexes or modulate their intracellular distribution in planta.Accordingly,it has been shown that a BiFC complex consisting of the ATF2-Jun heterodimer is translocated into the nucleus of mammalian COS cells after expression of the stress-activated p38SAPK protein kinase(van Dam et al.,1995;Hu et al.,2002).From the studies in mammalian cells using a set of different Jun-YFP N and Fos-YFP C fusion polypeptides,it has been estimated thatfluorescence complementation may occur when the YFP fragments are separated by an average distance of greater than100A˚(Hu et al.,2002).A prerequisite for complementation over long distance is sufficientflexibility of the protein backbone,to which the fragments are fused.This then allows association of the YFP fragments to stabilize the initial BiFC complex.There-fore,the dynamics of BiFC complex formation in combi-nation with sufficient complementation over long distance may enable the identification of novel protein interactions in plant cells by genetic screens based,for example,on high-efficiency protoplast transformation accompanied by fluorescence-activated cell sorting(Birnbaum et al.,2003; Galbraith,2004).Furthermore,by generating large popu-lations of transgenic plants expressing distinct YFP N or YFP C fusion proteins,interaction screens can also be performed on whole plant level by crossing individual lines.Protein interaction in plant cells435ªBlackwell Publishing Ltd,The Plant Journal,(2004),40,428–438。
介绍折纸的英文作文六年级1. Folding paper is a fun and creative activity that people of all ages can enjoy. It is a great way to relax and express your creativity. You can fold paper into various shapes and objects, such as animals, flowers, and even buildings. It is amazing how a simple piece of paper can be transformed into something beautiful and three-dimensional.2. When folding paper, it is important to have a steady hand and pay attention to the details. Each fold must be precise and accurate in order to achieve the desired shape. It requires patience and concentration, but the end result is always worth it. The satisfaction of seeing your creation come to life is truly rewarding.3. There are different types of paper that can be used for folding. Origami paper, for example, is specially designed for this purpose. It is thin and easy to fold, making it ideal for intricate designs. However, you canalso use regular paper or even recycled paper. The choice of paper depends on the complexity of the design and personal preference.4. Origami, the art of paper folding, originated in Japan. It has a long history and is deeply rooted in Japanese culture. In Japan, origami is not only a form of art but also a way to teach patience, concentration, and appreciation for simplicity. It is often used in educational settings to teach children about geometry and spatial reasoning.5. Folding paper can be a social activity as well. You can fold paper with your friends or family members and exchange your creations. It is a great way to bond and share your creativity with others. You can also join origami clubs or attend workshops to learn new techniques and meet fellow paper folding enthusiasts.6. The possibilities are endless when it comes to folding paper. You can create your own designs or follow instructions from books or online tutorials. There arecountless patterns and techniques to explore. Whether you are a beginner or an experienced folder, there is always something new to learn and discover.7. In conclusion, folding paper is a versatile and enjoyable activity that allows you to unleash your creativity. It is a form of art that can be practiced by anyone, regardless of age or skill level. So grab a piece of paper and start folding! You never know what amazing creations you can make.。
如何学好英语生物地理Learning a subject as complex and diverse as English Biology Geography can be a daunting task, but with the right approach and dedication, it is certainly achievable. In this guide, we will outline some key strategies and tips to help you master this fascinating field of study.1. Develop a Strong Foundation in EnglishBefore diving into the more intricate aspects of Biology Geography, it is important to have a solid grasp of the English language. This includes not only being able to read and comprehend English texts, but also to effectively communicate your ideas in writing and through spoken language.To improve your English skills, consider taking an English language course or enrolling in a language exchange program. Practice reading English books, articles, and news stories on topics related to biology and geography to enhance your vocabulary and comprehension. Additionally, watch English documentaries and videos on relevant subjects to become more familiar with scientific terminology and concepts.2. Familiarize Yourself with Biological ConceptsOne of the key components of English Biology Geography is understanding basic biological concepts and principles. Start by familiarizing yourself with the structure and function of cells, organisms, and ecosystems. Learn about genetics, evolution, and environmental science to develop a comprehensive understanding of the natural world.To facilitate your learning, consider using textbooks, online resources, and interactive learning tools such as videos and quizzes. Create study guides and flashcards to help you remember important terms and concepts. Make connections between different biological topics to gain a deeper understanding of how living organisms interact with their environment.3. Explore Geographic Principles and PracticesIn addition to mastering biological concepts, it is essential to understand the geographical aspects of English Biology Geography. Learn about the Earth's physical features, climate patterns, and ecosystems to appreciate the interconnectedness of life on our planet. Study maps, graphs, and charts to analyze spatial relationships and patterns in the natural world. To enhance your geographic knowledge, consider taking field trips to local parks, nature reserves, and other natural environments. Observe the unique flora and fauna in different regions and note how they have adapted to their surroundings. Use geographic information systems (GIS) software to map and analyze data related to biological and environmental phenomena.4. Engage with Current Research and DevelopmentsStay informed about the latest research and developments in the fields of biology and geography to deepen your understanding of English Biology Geography. Follow scientific journals, websites, and social media channels to learn about groundbreaking discoveries and advancements in these disciplines. Attend lectures, conferences, and workshops to interact with experts and fellow enthusiasts in the field.Consider conducting your own research projects or experiments to apply your knowledge and skills in a practical setting. Collaborate with classmates, teachers, or mentors to explore topics of interest and contribute to the scientific community. Develop critical thinking and problem-solving skills by analyzing data, drawing conclusions, and presenting your findings in a clear and concise manner.5. Practice Effective Study Habits and Time ManagementTo succeed in learning English Biology Geography, it is important to cultivate effective study habits and time management skills. Create a study schedule that allows you to dedicate regular time to review course materials, complete assignments, and prepare for exams. Break down complex topics into smaller, more manageable units to avoid feeling overwhelmed.Use a variety of study techniques, such as note-taking, highlighting, and summarizing, to reinforce your understanding of key concepts. Review your notes and textbooks regularly to retain information and consolidate your knowledge. Seek feedback from teachers, tutors, or peers to identify areas for improvement and adjust your study strategies accordingly.6. Seek Support and ResourcesDon't be afraid to reach out for help and support when needed while learning English Biology Geography. Consult with your teachers, advisors, or classmates if you have questions or need guidance on specific topics. Join study groups or online forums to collaborate with like-minded students and share resources and strategies.Utilize academic support services, such as tutoring centers or writing labs, to enhance your writing, research, and analytical skills. Take advantage of library resources, online databases, and scientific journals to access a wide range of information and data on biological and geographical topics. Stay motivated and focused on your goals by surrounding yourself with a supportive and encouraging community of learners.In conclusion, learning English Biology Geography requires dedication, perseverance, and a willingness to explore and understand the natural world. By developing a strong foundation in English, familiarizing yourself with biological concepts, exploring geographic principles, engaging with current research, practicing effective study habits, and seeking support and resources, you can enhance your knowledge and skills in this fascinating field of study. Embrace the challenges and opportunities that come with mastering English Biology Geography, and enjoy the journey of discovery and exploration.。
多模态训练方法下载温馨提示:该文档是我店铺精心编制而成,希望大家下载以后,能够帮助大家解决实际的问题。
文档下载后可定制随意修改,请根据实际需要进行相应的调整和使用,谢谢!本店铺为大家提供各种各样类型的实用资料,如教育随笔、日记赏析、句子摘抄、古诗大全、经典美文、话题作文、工作总结、词语解析、文案摘录、其他资料等等,如想了解不同资料格式和写法,敬请关注!Download tips: This document is carefully compiled by the editor. I hope that after you download them, they can help you solve practical problems. The document can be customized and modified after downloading, please adjust and use it according to actual needs, thank you!In addition, our shop provides you with various types of practical materials, such as educational essays, diary appreciation, sentence excerpts, ancient poems, classic articles, topic composition, work summary, word parsing, copy excerpts, other materials and so on, want to know different data formats and writing methods, please pay attention!多模态训练方法在近年来备受关注,成为了学术界和产业界广泛探讨的热点话题。
初三英语知识点大总结语法Grammar1. Nouns: A noun is a word that represents a person, place, thing, or idea. Nouns can be singular or plural. Examples of nouns include "dog," "school," and "happiness."- Common nouns refer to general people, places, things, or ideas, while proper nouns referto specific ones. For example, "dog" is a common noun, while "Spot" is a proper noun.- Nouns can also be concrete (tangible) or abstract (intangible).2. Pronouns: Pronouns are words used to replace nouns in a sentence. They include words like "he," "she," "it," "they," and "we." An example of a pronoun in a sentence is "She is going to the store."3. Verbs: Verbs are words that show action or state of being. They are an essential part of every sentence. For example, "run," "swim," and "think" are all verbs.- Verbs can be regular or irregular, and they can also be transitive or intransitive.4. Adjectives: Adjectives are words that describe or modify nouns. They provide more information about the noun in a sentence. For example, "beautiful," "tall," and "blue" are all adjectives.- Adjectives can be used to compare things, such as in the case of comparative (e.g., "bigger") and superlative (e.g., "biggest") forms.5. Adverbs: Adverbs are words that modify verbs, adjectives, or other adverbs. They provide information about how, when, where, or to what extent an action is taking place. For example, "quickly," "slowly," and "well" are all adverbs.6. Prepositions: Prepositions are words that show the relationship between a noun or pronoun and other words in a sentence. They are often used to indicate location, time, or direction. Examples of prepositions include "in," "on," "by," and "under."7. Conjunctions: Conjunctions are words that connect words, phrases, or clauses in a sentence. Examples of conjunctions include "and," "but," "or," and "so." They are used to join ideas together and make the structure of a sentence more complex.8. Articles: Articles are words used to specify or identify a noun. There are two types of articles: "the" (definite article) and "a" or "an" (indefinite articles). For example, "the dog" refers to a specific dog, while "a dog" refers to any dog.9. Tenses: Tenses refer to the time of an action or state of being in a sentence. The three primary tenses in English are past, present, and future. Each tense can be further divided into simple, continuous (progressive), perfect, and perfect continuous forms.10. Passive voice: The passive voice is used to focus on the action rather than the person or thing performing the action. In a passive sentence, the object of the action becomes the subject of the sentence. For example, "The book was read by the student" is a passive construction, while "The student read the book" is active.Vocabulary1. Synonyms and antonyms: Synonyms are words that have similar meanings, while antonyms are words that have opposite meanings. It is important to build a strong vocabulary by learning synonyms and antonyms of common words.2. Homophones and homographs: Homophones are words that sound the same but have different meanings and spellings (e.g., "to," "two," and "too"). Homographs are words that are spelled the same but have different meanings and pronunciations (e.g., "tear" as in "tear the paper" and "tear" as in "tear drop").3. Word families: A word family consists of a base word and its related forms. For example, the word family for the base word "act" includes "act," "active," "actor," and "action."4. Idioms and phrasal verbs: Idioms are figures of speech with meanings that are different from the literal meanings of the words used. Phrasal verbs are combinations of a verb and one or more particles (e.g., "in," "on," "up") that have idiomatic meanings. For example, "break up" and "get over" are both phrasal verbs.5. Collocations: Collocations are words that are frequently used together. For example, we say "have a shower" instead of "take a shower," and "strong coffee" instead of "powerful coffee."Reading comprehension1. Main idea: The main idea of a passage or text is the most important point that the author is trying to convey. It is crucial to be able to identify the main idea in a given passage to understand the overall message.2. Inference: Inference involves using clues from the text to make educated guesses about what is not explicitly stated. It requires critical thinking and an understanding of context.3. Context clues: Context clues are hints found within a sentence, paragraph, or passage that help the reader understand the meaning of unfamiliar words.4. Summarizing: Summarizing is the act of briefly retelling the main points of a passage in one's own words. It requires identifying the most important information and leaving out unnecessary details.5. Text structure: Text structure refers to how the information within a passage is organized. Common types of text structures include cause and effect, problem and solution, sequence, and compare and contrast.Writing1. Paragraph structure: A well-organized paragraph typically consists of a topic sentence, supporting sentences, and a concluding sentence.2. Punctuation: It is important to understand how to use various types of punctuation, including periods, commas, semicolons, and apostrophes, to ensure clear and effective writing.3. Spelling and grammar: Good spelling and grammar are essential for clear and effective communication. Proofreading is crucial to catch any errors before submitting a piece of writing.4. Descriptive writing: Descriptive writing uses sensory details to create a vivid picture of a person, place, or thing. It involves using adjectives and figurative language to paint a scene for the reader.5. Expository writing: Expository writing explains or informs. It is organized by topic and typically uses facts, examples, and explanations to convey information.6. Persuasive writing: Persuasive writing aims to convince the reader to embrace a particular point of view or take a specific action. It often includes persuasive techniques such as testimonials, emotional appeals, and logical reasoning.Listening and speaking1. Active listening: Active listening involves fully concentrating, understanding, responding, and remembering what is being said. It is an essential skill for effective communication.2. Speaking confidently: Confidence in speaking is developed through practice and a strong understanding of the language. It is important to speak clearly and fluently, use appropriate tone and body language, and engage with the audience.3. Giving and following directions: Giving and following directions requires clear and concise language, as well as an understanding of spatial relationships and directional vocabulary.4. Expressing opinions and asking questions: Being able to express opinions and ask questions in a conversation or discussion is essential for effective communication and building relationships.Overall, by mastering these essential grammar points, vocabulary, reading comprehension, writing, listening, and speaking skills, students in the 9th grade will be better equipped to communicate effectively and confidently in English. Practicing these skills regularly through reading, writing, listening, and speaking activities will further strengthen their language abilities and provide a solid foundation for success in their academic and professional endeavors.。
CS-TR-4730June2005Spatial Join Techniques∗Edwin H.Jacox and Hanan SametComputer Science DepartmentCenter for Automation ResearchInstitute for Advanced Computer StudiesUniversity of MarylandCollege Park,Maryland20742jacox@ and hjs@Keywords:spatial join,plane-sweep,external memory algorithms,spatial databasesAbstractA variety of techniques for performing a spatial join are presented.Rather thanjust summarize the literature,this in-depth survey and analysis of spatial join algo-rithms describes distinct components of the spatial join techniques,and decomposeseach algorithm from the literature into this framework.A typical spatial join articlewill describe many components of a spatial join algorithm,such as partitioning thedata,performing internal memory spatial joins on subsets of the data,and checkingif the full polygons intersect.Rather than describe a technique in its entirety,eachtechnique is decomposed and each component is addressed in a separate section so asto compare and contrast the similar pieces of each technique.The goal of this articleis to describe algorithms within each component in detail,comparing and contrastingcompeting methods,thereby enabling further analysis and experimentation with eachcomponent and allowing for the best algorithms for a particular situation to be builtpiecemeal,or even better,allowing an optimizer to choose which algorithms to use.Table1:Spatial join components.Section3Internal Memory Methods3.2Index Nested-Loop Join[28]3.4Z-Order[5,77]4.1.1Hierarchical Traversal[20,36,48,53] Both Sets Indexed4.1.3Multi-Dimensional Point Methods[96]4.2.1Construct a Second Index[60]One Data Set Not Indexed4.2.3The Index as Sorted Data[8,38]4.3.1External Plane Sweep[50]Neither Set Indexed4.3.6Grid Partitioning[88,106]4.3.8Size Partitioning[56,9]Section6Refinement6.2Polygon Intersection Test[91,19]Table2:Spatial join issues.Section2Spatial Join Basics2.2Minimum Bounding Rectangles Processing IssuesPartitioning Issues4.3.5Avoiding Duplicate ResultsSection5Alternate Filtering Techniques5.2True Hit Filtering[16]Section7.4Selectivity EstimationNon-Uniform Data Set Estimates[14,30,68]Table3:Specialized spatial joins.Section7.1Multiway Spatial Joins7.1.2Multiway Hierarchical Traversal[70,67,83] Section7.2Parallel Spatial JoinsParallel Grid Partitioning Methods[64,89,106] Section7.3Distributed Spatial Joins1.The data sets can be objects other than rectangles such as points,lines,or polygons.2.The data sets might have more than two dimensions.3.The relationship between pairs of objects can be any relation between the objects,such as intersection,nearness,enclosure,or a directional relation(e.g.,find all pairs of objects such that r is northwest of s[109]).4.There might be more than two data sets in the relation(a multiway spatial join)oronly one set(a self spatial join).Spatial joins are distinguished from a standard relational join[72]in that the join con-dition involves the multi-dimensional spatial attribute of the joined relation.This property prevents the use of the more sophisticated relational join algorithms.For instance,because the data objects are multi-dimensional,there is no ordering of the data that preserves prox-imity.Relational join techniques that rely on sorting the data,such as the sort-merge join [72],work because neighboring objects(those with the next higher and lower value)are adjacent to each other in the ordering.However,in more than one dimension,the data can not be sorted so that this property holds.For example,in two dimensions,the left and right neighbors can be adjacent to an object in an ordering,but then the top and bottom neighbors will need to go elsewhere in the order(see Section2.3for a further discussion of multi-dimensional orderings).Other relational join techniques are also inapplicable because the data objects might have extent.For example,equijoin techniques[72](e.g.hash joins),will not work with spatial data because they rely on grouping objects with the same value,which is not possible when the objects have extent.This is the same reason that the techniques will not work with intervals(extent in one dimension)or inequalities.As an example,for a one-dimensional hash join on sets R and S,a group of objects,say G,is formed from set R and placed in a bucket.If any object g in bucket G satisfies the relation with an object from set S,then so does every object from set G.This property does not hold for objects with extent,such as a rectangle,because the objects can overlap each other and a disjoint grouping might not exist.In fact,an object from set R could potentially intersect every object from set S. Because of these two factors,not proximity preserving and extent,relational join algorithms can not be used directly to perform a spatial join.The computational geometry approach to solving the simplified spatial join(a two-set rectangle intersection)is to use a plane-sweep technique[91](see Section3.3).In order to use the plane-sweep method for a general spatial join,two problems must be overcome:that the objects are not rectangles and that there might be insufficient internal memory for the plane-sweep algorithm.Furthermore,calculating whether two complex objects satisfy the join condition,such as intersection,can be an expensive operation,and performing as few of these operations as possible improves overall performance.To overcome these problems, a spatial join is typically performed in a two stagefilter-and-refine approach[79].In thefilter-and-refine approach,the spatial join isfirst solved using approximations of the objects in thefiltering stage and any incorrect results due to the approximations are removed in the refinement stage using the full objects1.In thefiltering stage,objects aretypically approximated using minimum bounding rectangles(see Section2.2),hereafter re-ferred to as MBR’s,which require less storage space than the full object,making processing and I/O operations less expensive2.For example,GIS objects might be polygons,each consisting of thousand of points.Reading these objects in and out of memory could easily be the dominant cost of performing a spatial join,whereas afilter-and-refine approach al-leviates this problem.Furthermore,a spatial join on rectangles presents a more tractable problem.For smaller data sets,thefiltering stage of the spatial join can be solved using internal memory techniques,which are described in Section3.For larger data sets,external (secondary)memory techniques are required for thefiltering stage,which are described in Section4.The output of thefiltering stage is a list of all pairs of objects whose approximations satisfy the join condition,which is referred to as the candidate set,and is typically rep-resented by pairs of object ids.The candidate set includes all of the desired pairs,those whose full objects intersect,but also includes pairs whose approximations satisfy the join condition,but whose full objects do not.The extra pairs appear because of the inaccuracy of the object approximations(see Section2.2).The purpose of the refinement stage is to remove the undesired pairs using the full objects,producing thefinal list of object pairs that satisfy the given join condition.Refinement techniques are described in Section6.As mentioned,the dominant cost of a spatial join with very large objects is the I/O cost of reading the large objects.Earlyfiltering techniques were dominated by I/O costs[20]. Later techniques have improved I/O performance so that it is no longer an axiom that I/O costs dominate the CPU costs[88].Even thoughfiltering reduces the I/O costs,reading large objects can still be the major cost of the refinement stage,which is generally more expensive than thefiltering stage[88].Furthermore,while the performance improvement from using afilter-and-refine approach might be obvious for very large objects,it remains an open question as to whether it is the best approach for smaller,simpler objects.As an example of an alternative approach,Zhu et al.[108,107]have proposed methods for extending the plane-sweep algorithm(Section3.3)to trapezoids and recti-linear polygons, thereby avoiding the need for thefilter-and-refine approach.Throughout the review of techniques,we avoid discussing experimental results.Most of the methods in this article were shown to outperform some other method.Wefind most of these experimental results to be inconclusive because they frequently use only a few data objects,the techniques are compared with one or no other technique,and the implementations of the techniques can vary dramatically,which has a large impact on results.Furthermore,the variety of computer hardware,software and networks used make it difficult to compare results between methods.For these reasons,we do not discuss most experimental results.Nevertheless,we do point out the data set generator of G¨u nther et al.[37],which helps to establish a benchmark for spatial joins beyond the typical Sequoia [97]and Tiger[76]data sets.Benchmarks are an important step towards achieving the goal of repeatable and predictable algorithm performance.Also,to simplify the discussion of the techniques,it is assumed that the data is two dimensional and that we are interested in determining pairs of intersecting objects.Both of these assumptions are common in the literature.The two-dimensional assumption is made because higher dimensional data has not been addressed in the literature in regards to spatial joins and many of the techniques presented might not work or might not perform well in higher dimensions.The intersection assumption is made only to simplify the discussionand believe that this assumption does not effect the generality of the algorithms.For example,a nearness relation can easily be calculated by extending the size of the MBR’s so that nearness is calculated by an intersection test[57].When appropriate,a generic join condition is used,rather than intersection.Furthermore,although many spatial join techniques depend on spatial indices,the dis-cussion of spatial indices is left to other work[34,94].Knowledge of these structures can be crucial to a deeper understanding of many of the techniques for processing spatial data. Where appropriate,these index structures are described,but mostly the algorithms are presented in such a way that little or no knowledge of the underlying spatial indices is required.The remainder of this section discusses issues that are fundamental to the design of spatial join algorithms.Section2.1discusses various design considerations and parameters that influence the performance of spatial join algorithms.Sections2.2and2.3review MBR’s and linear orderings,respectively,which are concepts that are fundamental to many spatial join techniques.2.1Design Considerations and ParametersMany factors contribute to the performance of a spatial join and influence the design of al-gorithms.The foremost factor,of course,is the processor speed and I/O performance.More importantly for the design is the ratio of these two factors.Early spatial joins algorithms were constrained by I/O,which dominated CPU time,and the focus of improvements was on minimizing the amount of data that needed to be read from and written to external mem-ory.As spatial join algorithms improved,experiments showed that CPU time accounted for an equal share of performance and that the algorithms were no longer I/O dominated[20]. Today,algorithms need to account for both CPU performance and I/O performance.These two factors can be balanced somewhat by tuning page sizes and buffer sizes(the amount of internal memory available to the algorithm),two factors which also play an important role in performance.However,as processor,I/O speeds,and internal memory sizes continue to improve,algorithms need to account for these factors and thus,tuning will always be necessary for the best performance.The characteristics of the data sets and whether the data sets are indexed are also major influences on performance.The data set sizes obviously effect overall performance,but more importantly is whether or not the data setfits into the available internal memory.If the entire data set doesfit in internal memory,then the spatial join can be done entirely in memory(Section3),which can be significantly faster than using external memory methods (Section4).One of the most confounding factors for spatial join design is the distribution of data.Algorithms for uniformly distributed data sets are easy to develop,but algorithms for handling skewed data sets are significantly more complicated.A poorly designed algorithm can thrash with skewed data sets,repeatedly reading the same data in and out of external memory,which severely degrades performance.These factors are mitigated if the data is indexed appropriately.If a data set is indexed,then generally,algorithms that use the index will be faster than those that don’t.Section4classifies spatial join algorithms by whether they assume that both data sets are indexed(Section4.1),only one data set is indexed (Section4.2),or neither of the data sets are indexed(Section4.3).How the data is stored is another factor that contributes to the design of spatial joins. Vectors(a list of vertices)are commonly used to store polygons,but raster approaches are also used[77].The choice of storage method for the full object mostly effects the complexityof the object intersection test during the refinement stage,since an approximation of the full object is used during thefiltering stage.This article only discusses refinement techniques for the more common vector representation.During thefiltering stage,an object is represented by an approximation and an object id or a pointer is used to access the full object.An MBR is generally chosen as the approximation,but other approximations can also be used (see Section5).The environment in which the algorithm runs is also a consideration in the design of spatial join algorithms.In a pipe-lined system[35],for instance,each stage of the spatial join algorithm needs to output results continuously in order for the pipe line to run efficiently. In this case,each stage is said to be non-blocking because the next stage does not need to wait for results.Unfortunately,filtering methods that sort or partition the data are blocking,although there is a method to produce some results earlier(see Section5.3).Also, specialized algorithms can be used to improve the performance of multiway spatial joins,and modified algorithms are required to perform spatial joins that run in parallel environments and distributed environments(see Section7).2.2Minimum Bounding Rectangles and Approximations(a)(b)(c)Figure1:(a)A minimum bounding rectangle(MBR)is the smallest rectangle that fully encloses an object and whose sides are parallel to the axes.(b)The area of an MBR might be significantly larger than the area of the enclosed object.The extra area is referred to as dead space.(c)Two MBR’s might intersect even though the objects that they enclose do not intersect.For thefiltering stage,most algorithms use a minimum bounding rectangle(MBR)to approximate the full object,though other approximations might be used instead or as a secondaryfilter(Section5).An MBR of an object is the smallest enclosing rectangle whose sides are parallel to the axes of the space(iso-oriented),as shown in Figure1a.MBR’s are preferred over the full object because they require less memory and intersections between MBR’s are easier to calculate.Objects,especially in GIS applications,can be very large, requiring many points or lines to represent a polygon,and for large data sets,which are also typical in GIS applications,reading thousands,millions,or more of these objects from external memory and performing intersection tests on them can be extremely expensive in terms of I/O performance.Instead,if MBR’s are used in thefiltering stage,the MBR’s can be read from external memory faster then the full object and the intersection tests performed faster.Unfortunately,using MBR’s,or any approximation,will produce somewrong answers.As shown in Figure1b,an object might only occupy a fraction of its MBR, leaving a portion of dead space.Two MBR’s might intersect,but the objects they represent might not intersect,as shown in Figure1c.This result is referred to as a false hit,whereas the result is termed a true hit if the MBR’s intersect and the objects they represent also intersect.Before performing a spatial join,the MBR’s for the full objects must be calculated.If the data is indexed using a spatial indexing method[34,94],then typically the MBR’s exist already.If they do not,then a scan of the full data set is required to create the MBR’s. Forming the MBR of an object simply involves checking each corner point of the object, which is an O(n)operation for a polygonal object with n vertices.(a)(b)Figure2:(a)In order to reduce dead space,an object can be approximated by two disjoint rectangles.(b)However,both rectangles might intersect a second object,thereby producing duplicate results.Use of thefilter and refine approach for spatial joins wasfirst introduced by Orenstein [79].Orenstein was concerned that a poor approximation would degrade performance[78] and experimented with using a set of disjoint rectangles to approximate each object.For example,to use this representation,the object in Figure1b is decomposed into the two MBR’s shown in Figure2a,improving the approximation by reducing the dead space,but also increasing the size of the data set.While this approach improves the accuracy of the filter stage,it also creates the need for an extra step after thefiltering stage to remove duplicates from the candidate set.As shown in Figure2,both pieces of the decomposed object in Figure2a might intersect the same object,as shown in Figure2b.Both of these intersections create a candidate pair.The duplicate results generally need to be removed for most applications and this typically should be done before the more costly refinement stage in order to avoid extra processing(see Section4.3.5for a discussion of duplicate removal techniques).Nevertheless,most algorithms use one MBR,rather than approximating an object by a set of rectangles,and rely on the refinement stage to efficiently remove false hits.However, many algorithms intentionally duplicate objects.For instance,if an algorithm creates a disjoint partition of the objects in a divide-and-conquer approach,as is done with a grid partitioning approach(see Section4.3.6),then each object will appear in each partition it overlaps.Similarly,some spatial indices that can be used to perform a spatial join use disjoint nodes and the objects again are copied into each node they overlap(e.g.,the R+-tree[95]).In both cases,duplicate removal(or avoidance)techniques are required(seeSection4.3.5).2.3Linear Orderings(a)(b)Figure3:(a)A linear order,where object a’s neighbors,b and c are nearby.(b)A linear order in which one of object a’s neighbors,c,is nearby,but the other neighbor,b,is not.A linear order[51,94]creates a total order on multi-dimensional objects.In other words, a linear order is a traversal of all of the objects,as shown in Figure3.Linear orderings play an important role in many spatial join techniques,in a similar way that sorted orders (a one-dimensional linear ordering)play an important role in creating efficient algorithms for relational joins(e.g.the sort-merge join[72]).The benefit of a sorted order in one dimension is that neighboring objects(close in value)are next to each other in the sorted order,leading to algorithms such as the sort-merge join[72].In more than one dimension, no natural linear order exists and spatially neighboring objects might not be close in the linear order.For example,in Figure3a,neighboring objects a and b are next to each other in the order indicated by the arrows,but in Figure3b,they are widely separated.Even so,because linear orders keep some of the neighboring objects near each other in the order, they can be useful in spatial join techniques.Linear orders that keep neighboring objects closer in the order,on average,such as the Z-order or the Peano-Hilbert order,tend to be more useful for spatial join algorithms.The Z-order(also known as a Peano or Morton order),shown in Figure4a,and the Peano-Hilbert order,shown in Figure4b,traverse the grid in a pattern that helps to preserve locality.Both of these linear ordersfirst order objects in a block before moving to the next block.For example,the Z-order is a traversal through a regular grid using a‘Z’pattern,as shown in Figure5a.If there are more than four cells in the grid,then each top-level block is fully traversed before moving to the next block.In Figure5b,block A from Figure5a is traversed in a‘Z’pattern before moving to block B.This pattern is repeated atfiner levels,where a block at any level is fully traversed before moving to the next block.In this way,a linear order is imposed on the cells.The Peano-Hilbert order is similar,as shown in Figure4b,though each block is traversed in a rotation that might be clockwise or counter-clockwise,avoiding the large jumps between the constituent grid cells of a Z-order.Additionally,an order might visit both the grid cells and the enclosing blocks,for instance,ordering both the blocks in Figure5a and the grid cells in Figure5b.To accomplish(a)(b)Figure4:(a)Z-Order(Peano order)and(b)Peano-Hilbert order.this order,one convention is to visit enclosing regions(the blocks in Figure5a)before visiting smaller regions(the cells in Figure5b),as shown in Figure5c,which also includes the top level cell(the enclosing space)in the ordering.In essence,this a hierarchical traversal of the nodes,as shown in Figure5d.To traverse points in a linear order,the grid cells can be made small enough such that each point is in its own grid cell.To traverse objects in a linear order,either a point on the objects,such as the centroid,is used to represent each object or the objects are assigned to the smallest enclosing block or grid cell,which is similar to creating an MBR for the object, but with more dead space,as shown in Figure6.Note that an object,no matter how small, that intersects the center point will always be in the top level cell(root space),as shown in Figure7.Some algorithms can take advantage of the regular structure of the enclosing cells,but at the price of more false hits due to the increased dead space(see Sections3.4 and4.3.8).3The Filtering Stage–Internal MemoryDuring thefiltering stage,a spatial join is performed on approximations of the objects.This section describes techniques for performing a spatial join without using external memory, that is,no data is written to external memory.If there is insufficient internal memory to process a spatial join entirely in memory,then external memory must be used to store all or portions of the data sets during processing(see Section4).Even so,at some point,most external memory spatial join algorithms reduce the size of the problem and process subsets of the data using internal memory techniques.Section3.1first describes the brute force nested-loop join.Next,Section3.2describes the related index nested-loop join,which is presented as an internal memory method even though it can be used as an external memory algorithm if the indices are stored in external memory.Two more sophisticated approaches are also described:the plane-sweep algorithm, rooted in computational geometry,in Section3.3,and a variant of the plane-sweep that uses a linear ordering of the data,in Section3.4.(a)BD(b)D(c)(d)Figure5:(a)A four cell grid traversed in a Z-order.(b)A sixteen cell grid traversed ina Z-order.(c)A sixteen cell grid traversal that includes enclosing blocks(lettered blocks).(d)A sixteen cell grid traversal as a tree traversal.3.1Nested Loop JoinsThe most basic spatial join method is the nested-loop join,which compares every object inone set to every object in the other set[72].The algorithm,shown in Figure8,takes everypossible pair of objects and passes it to the SATISFY function to check if the pair of objectsmeets the given join condition,termed joinCondition in the algorithm.If a pair satisfiesthe join condition,then it is reported using the REPORT function.Given two data sets,Aand B,with n a and n b objects in each,respectively,the nested-loop join takes O(n a·n b) time.Despite this larger cost,the nested-loop join can be useful when there are too fewobjects to justify the overhead of more complex methods.Note that this algorithm workswith any object type and with any arbitrary join condition.3.2Index Nested-Loop JoinA variant of the nested-loop join algorithm given in Section3.1,called the index nested-loopjoin[28],improves performance for larger data sets byfirst creating a spatial index on oneset,say A.In this algorithm,given in Figure9,the spatial index isfirst created and everyelement of set A is inserted into the index using the INSERT function.Then,the other set,say B,is scanned,and each element is used to search the index on set A for intersections.The index is searched using the SEARCH function,which in this context becomes a windowquery[34]on the index,where the window is the object from set B.Generally,the searchwindow is a rectangle,which limits the types of join conditions to intersection tests orrelated relations that can be solved with a window query,such as proximity.Typically, the time to search the index is O(log(n a)+f),where n a is the size of set A and f is theA BD CFigure 6:In a linear ordering,objects can be assigned to the smallest enclosing block or grid cell.Object r is assigned to the root space since it is not within any block.Object s is assigned to the lower right block,B,and object t is assigned to cell14.Figure 7:An object overlapping the center point,no matter how small,will be assigned to the top level,which is the entire space.number of intersections found.In theory,an object could intersect every object in the index,creating an O (n )search time.In practice though,the number of intersections is small and the running time of the entire algorithm is O ((n a +n b )·log (n a )+f ),which includes the time to construct the index,which is typically O (n a ·log (n a )).Since all of set A is inserted first,more efficient static indices and bulk-loading techniques [45,101]can be used to improve the construction time and the performance of the index.The index nested-loop algorithm can be executed entirely in memory,using in memory indices,and is useful as a component in other spatial join algorithms (see Section 4).The algorithm can also be used as a stand alone external memory spatial join algorithm by using external memory indices,which allows the algorithm to process larger data sets.For instance,Becker et al.[13]used grid files [75]as the index and Henrich and M¨o ller [43]used an LSD tree [44]as the index.However,more sophisticated methods exist for using an external spatial index to perform a spatial join (see Section 4).procedure NESTED_LOOP_JOIN(setA,setB,joinCondition);beginfor each a∈setAfor each b∈SetBif SATISFIED(a,b,joinCondition)thenREPORT(a,b);end if;next b;next a;end;Figure8:The basic nested loop join with running time O(n a·n b),for data sets of size n a and n b.procedure INDEX_NESTED_LOOP_JOIN(setA,setB);beginspatialIndex←CREATE_SPATIAL_INDEX(setA);for each a∈setAspatialIndex.INSERT(a)end for each;for each b∈setBsearchResults←spatialIndex.SEARCH(b)REPORT(searchResults)end for each;end;Figure9:An index nested-loop join improves the performance of the spatial join to O((n a+ n b)·log(n a)+f),assuming search times of the index are O(log(n a)+f),where f is the number of intersections found,n a is the size of the indexed data set,and n b is the size of the unindexed data set.3.3Plane SweepA two-dimensional plane-sweep[91]of a set of iso-oriented rectanglesfinds all of the rect-angles that intersect.The algorithm has two passes.Thefirst pass sorts the rectangles in ascending order on the basis of their left sides(i.e.,x coordinate values)and forms a list. The second pass sweeps a vertical scan line through the sorted list from left to right,halting at each one of these points,say p.At any instant,all rectangles that intersect the scan line are considered active and are the only ones whose intersection needs to be checked with the rectangle associated with p.This means that each time the sweep line halts,a rectangle becomes active,causing it to be inserted into the set of active rectangles,and any rectangles entirely to the left of the scan line are removed from the set of active rectangles3.Thus, the key to the algorithm is its ability to keep track of the active rectangles(actually,just their vertical sides),as well as performing the actual intersection test.To keep track of the active rectangles,the plane-sweep algorithm uses a structure(re-。