R语言画图

来源：互联网发布：音乐剪切软件编辑：程序博客网时间：2024/05/17 22:22

一，布局

Ｒ绘图所占的区域，被分成两大部分，一是外围边距，一是绘图区域。

外围边距可使用par()函数中的oma来进行设置。比如oma=c(4,3,2,1)，就是指外围边距分别为下边距：4行，左边距3行，上边距2行，右边距1行。很明显这个设置顺序是从x轴开始顺时针方向。这里的行是指可以显示1行普通字体。所以当我们使用mtext中的line参数时，设置的大小就应该是[0,行数)的开区间。当我们使用mtext在外围边距上书写内容时，设置mtext中的outer=TRUE即可。

绘图区域可使用par()函数中的mfrow, mfcol来进行布局。mfrow和mfcol可以使用绘图区域被区分为多个区域。默认值为mfrow(1,1)。

比如mfrow(2,3)就是指将绘图区域分成2行3列，并按行的顺序依次绘图填充；
比如mfcol(3,2)就是指将绘图区域分成3行2列，并按列的顺序依次绘图填充；

我们将每一个细分的绘图区域分为两个部分，一是绘图边距，一是主绘图。

绘图边距需要容纳的内容有坐标轴，坐标轴标签，标题。通常来讲，我们都只需要一个x轴，一个y轴，所以在设置时，一般的下边距和左边距都会大一些。如果多个x轴或者y轴，才考虑将上边距或者右边距放大一些。绘图边距可以使用par()函数中mar来设置。比如mar=c(4,3,2,1)，与外围边距的设置类似，是指绘图边距分别为下边距：4行，左边距3行，上边距2行，右边距1行。很明显这个设置顺序是从x轴开始顺时针方向。行的概念与之前的相同。也可以使用mai来设置。mai与mar唯一不同之处在于mai不是以行为单位，而是以inch为单位。

SOUTH<-1; WEST<-2; NORTH<-3; EAST<-4; GenericFigure <- function(ID, size1, size2){  plot(0:10, 0:10, type="n", xlab="X", ylab="Y")  text(5,5, ID, col="red", cex=size1)  box("plot", col="red")  mtext(paste("cex",size2,sep=""), SOUTH, line=3, adj=1.0, cex=size2, col="blue")  title(paste("title",ID,sep=""))} MultipleFigures <- function(){  GenericFigure("1", 3, 0.5)  box("figure", lty="dotted", col="blue")   GenericFigure("2", 3, 1)  box("figure", lty="dotted", col="blue")   GenericFigure("3", 3, 1.5)  box("figure", lty="dotted", col="blue")   GenericFigure("4", 3, 2)  box("figure", lty="dotted", col="blue")} par(mfrow=c(2,2),mar=c(6,4,2,1),oma=c(4,3,2,1)) MultipleFigures() box("inner", lty="dotted", col="green")box("outer", lty="solid", col="green") mtext("Outer Margin Area (oma) of South: 6", SOUTH, line=1, cex=1, outer=TRUE) plotline<-function(n,direc){  for(i in 0:n){    mtext(paste("line",i,sep=""), direc, line=i, cex=1, col="black", adj=1, outer=TRUE)  }}plotline(4,SOUTH)

Ｒ绘图布局

但是，使用mfrow,mfcol只能是矩阵似的布局，如果我们需要简单地实际不规则的布局，那该怎么办呢？还有强大的layout()专门用于布局。

layout(mat, widths = rep(1, ncol(mat)), heights = rep(1,nrow(mat)),respect = FALSE)

其中，mat就是一个距阵，假设我们要画N个图，那么，mat就要描述每个图所在的位置，其中1…N-1都必须至少出现过一次。比如有三个图，我们希望的布局是第一排有一个图，第二排有两个图，那么mat<-matrix(c(1,1,2,3), nrow=2, ncol=2, byrow = TRUE)；如果希望第一排有两个图，第二排有一个图，那么就是mat<-matrix(c(1,2,3,3),nrow=2,ncol=2,byrow=TRUE)。很明显，还是将画布分成许多小格矩阵，这里就是2Ｘ2的，如果希望第1个图放入第一排的1，2格，那就在矩阵的第1，2的位置写入1，如果是希望第2个图放在第一排的1，2格，那就在矩阵的第1，2的位置写入2。

> attach(mtcars)> layout(matrix(c(1,1,2,3), 2, 2, byrow = TRUE))> hist(wt)> hist(mpg)> hist(disp)

layout不规则布局0

那么参数中的widths和heights又是做什么用的呢？它们就是用来指定每行或者每列的宽度和高度的。我们可以做以下的比较：

> attach(mtcars)> layout(matrix(c(1,1,1,1,2,2,2,3,2,2,2,3),3,4,byrow=TRUE))> hist(wt)> hist(mpg)> hist(disp)> layout(matrix(c(1,1,2,3), 2, 2, byrow = TRUE), +    widths=c(3,1), heights=c(1,2))> hist(wt)> hist(mpg)> hist(disp)

layout不规则布局1

我们发现，这两种方法得到的是相同的结果，在代码的书写上，第2种方法要更清晰一些。

想象一下，似乎每个图的位置必须是连在一起的小矩阵，如果，不构成连在一起小矩阵，会成为怎么样的一个结果呢？

> attach(mtcars)> layout(matrix(c(1,1,2,1,1,1),nrow=2,ncol=3,byrow=T))> hist(wt)> hist(mpg)

layout内嵌图布局

如果需要大图，小图，内嵌图这layout()虽然很容易实现，但多少有点麻烦，不是那么随心所欲。还是回过头来使用par()吧。使用par()的中fig=在画布任意位置上画图。在使用fig参数时，需要把画布理解成左下角为坐标(0,0)，右上角为(1,1)的一个坐标系。fig=c(x1,x2,y1,y2)来设置该参，x1<x2,y1<y2，x1,y1定位绘图区的左下角，x2,y2定位绘图区的右上角。使用new=TRUE参数来确认是否在原画布上继续画，还重新在一张新画布上开始画。

> par(fig=c(0,0.85,0,0.85), new=F)> plot(mtcars$wt, mtcars$mpg, xlab="Miles Per Gallon",+   ylab="Car Weight")> par(fig=c(0,0.8,0.55,1), new=TRUE)> boxplot(mtcars$wt, horizontal=TRUE, axes=FALSE)> par(fig=c(0.65,1,0,0.8),new=TRUE)> boxplot(mtcars$mpg, axes=FALSE)> mtext("Enhanced Scatterplot", side=3, outer=TRUE, line=-3)> par(fig=c(0.4,0.75,0.4,0.7),new=TRUE,mar=c(2,2,0,0),mgp=c(1,.4,0),cex=1,cex.lab=0.7,cex.axis=0.7)> hist(mtcars$mpg, main="")

par实现内嵌图

二，颜色

计算机保存及还原颜色时有多种方案，较为常用的是两个，RGB和HSV。R预设了657种颜色，可以通过colors()函数调用（或者英式拼写colours()）。比如我们常用的红，绿，蓝，

> colors()[c(552,254,26)][1] "red"   "green" "blue"

我们可以使用grep来调取我们感兴趣的颜色，

> grep("red",colors())[1] 100 372 373 374 375 376 476 503 504 505 506 507 524 525 526 527 528 552 553[20] 554 555 556 641 642 643 644 645 > colors()[grep("red",colors())][1] "darkred" "indianred" "indianred1" "indianred2" [5] "indianred3" "indianred4" "mediumvioletred" "orangered" [9] "orangered1" "orangered2" "orangered3" "orangered4" [13] "palevioletred" "palevioletred1" "palevioletred2" "palevioletred3" [17] "palevioletred4" "red" "red1" "red2" [21] "red3" "red4" "violetred" "violetred1" [25] "violetred2" "violetred3" "violetred4" > colors()[grep("sky",colors())][1] "deepskyblue" "deepskyblue1" "deepskyblue2" "deepskyblue3" [5] "deepskyblue4" "lightskyblue" "lightskyblue1" "lightskyblue2"[9] "lightskyblue3" "lightskyblue4" "skyblue" "skyblue1" [13] "skyblue2" "skyblue3" "skyblue4" > SetTextContrastColor <- function(color)+ {+   ifelse( mean(col2rgb(color)) > 127, "black", "white")+ }> # Define this array of text contrast colors that correponds to each> # member of the colors() array.> TextContrastColor <- unlist( lapply(colors(), SetTextContrastColor) ) > # 1a. Plot matrix of R colors, in index order, 25 per row.> # This example plots each row of rectangles one at a time.> colCount <- 25 # number per row> rowCount <- 27> plot( c(1,colCount), c(0,rowCount), type="n", ylab="", xlab="",+   axes=FALSE, ylim=c(rowCount,0))> title("R colors")> > for (j in 0:(rowCount-1))+ {+   base <- j*colCount+   remaining <- length(colors()) - base+   RowSize <- ifelse(remaining < colCount, remaining, colCount)+   rect((1:RowSize)-0.5,j-0.5, (1:RowSize)+0.5,j+0.5,+     border="black",+     col=colors()[base + (1:RowSize)])+   text((1:RowSize), j, paste(base + (1:RowSize)), cex=0.7,+     col=TextContrastColor[base + (1:RowSize)])+ }

Ｒ颜色表

对于大多数理工出身的人来讲，理解颜色并不难，难的是如何选择一种或者多种理想的颜色，让绘图很漂亮。Ｒ当中有一个包RColorBrewer就可以为我们解决这个难题，其中预设了很多种颜色组合以供我们使用。

> library(RColorBrewer)> display.brewer.all()

RColorBrewer中预设的颜色表

我们看到其中Set3和Paired设定了12种颜色，其余的有多有少，以9种居多。在使用其颜色时，使用brewer.pal(n, name)调用即可，其中n最小值为3最大值为每组预设值数组的长度。比如brewer.pal(12,Paired)

在绘图时，有如下颜色参数

参数描述col绘图使用的颜色，许多函数接受一组颜色，并对不同的数据依次使用颜色。col.axis坐标轴字符颜色col.labx,y坐标标记颜色col.main标题颜色col.sub副标题颜色fg绘图前景色，包括坐标轴，各类boxesbg绘图背景色

> require(graphics)> par(col.axis="green",col.lab="blue",col.main="darkred",fg="white",bg="black")> plot(cars,main="speed vs dist")

颜色设置示例

三，字体

字体参数如下：

参数描述font字体描述，1正常，2加粗，3斜体，4加粗，斜体，5符号font.axis坐标轴字符描述font.lab坐标轴标记字体描述font.main标题字体描述font.sub副标题字体描述ps字体点阵大小，大约为1/72英寸。在使用时text size=ps*cexcex字体放大或者缩小多少倍cex.axis坐标轴字符大小cex.lab坐标轴标记字体大小cex.main标题字体大小cex.sub副标题字体大小family绘图字体。标准字体是”serif”,”sans”,”mono”,”symbol”。当然可以指定任何自己已有的字体库。但它是设备依赖的。

> fonts<-names(pdfFonts())> cnt<-length(fonts)> x<-cnt+10> x<-1:x> y<-x> pdf("font.pdf")> plot(x,y,type="n")> for(i in 1:cnt){text(i,i*1.3,family=fonts[i],paste("font",fonts[i]),adj=0)}> dev.off()

字体

四，符号与线形

画图中的符号由pch参数来控制。其描边色由col控制，填充色由bg控制。

> pchShow <-+   function(extras = c("*",".", "o","O","0","+","-","|","%","#"),+            cex = 3, ## good for both .Device=="postscript" and "x11"+            col = "red3", bg = "gold", coltext = "brown", cextext = 1.2,+            main = paste("plot symbols :  points (...  pch = *, cex =",+                         cex,")"))+   {+     nex <- length(extras)+     np  <- 26 + nex+     ipch <- 0:(np-1)+     k <- floor(sqrt(np))+     dd <- c(-1,1)/2+     rx <- dd + range(ix <- ipch %/% k)+     ry <- dd + range(iy <- 3 + (k-1)- ipch %% k)+     pch <- as.list(ipch) # list with integers & strings+     if(nex > 0) pch[26+ 1:nex] <- as.list(extras)+     plot(rx, ry, type="n", axes = FALSE, xlab = "", ylab = "",+          main = main)+     abline(v = ix, h = iy, col = "lightgray", lty = "dotted")+     for(i in 1:np) {+       pc <- pch[[i]]+       ## 'col' symbols with a 'bg'-colored interior (where available) :+       points(ix[i], iy[i], pch = pc, col = col, bg = bg, cex = cex)+       if(cextext > 0)+           text(ix[i] - 0.3, iy[i], pc, col = coltext, cex = cextext)+     }+   }> > pchShow()

符号

线形主要由lty和lwd来控制。lty: line type. lwd: line width.顾名思意，lty控制线的形状，而lwd控制线的粗细，默认值为1。设计成2表示两倍线宽。

线形

五，坐标轴，图例，标记与标题

六，各类图型

点，线，面

点：points，线：abline,lines,segments，面：box,polygon,polypath,rect，特殊的：arrows,symbols

points不仅仅可以画前文中pch所设定的任意一个符号，还可以以字符为符号。

> ## ------------ test code for various pch specifications -------------> # Try this in various font families (including Hershey)> # and locales.  Use sign=-1 asserts we want Latin-1.> # Standard cases in a MBCS locale will not plot the top half.> TestChars <- function(sign=1, font=1, ...)+ {+    if(font == 5) { sign <- 1; r <- c(32:126, 160:254)+    } else if (l10n_info()$MBCS) r <- 32:126 else r <- 32:255+    if (sign == -1) r <- c(32:126, 160:255)+    par(pty="s")+    plot(c(-1,16), c(-1,16), type="n", xlab="", ylab="",+         xaxs="i", yaxs="i")+    grid(17, 17, lty=1)+    for(i in r) try(points(i%%16, i%/%16, pch=sign*i, font=font,...))+ }> TestChars()> try(TestChars(sign=-1))

画点

abline可以由斜率和截距来确定一条直线，lines可以连接两个或者多个点，segments可以按起止位置画线。

> require(stats)> sale5 <- c(6, 4, 9, 7, 6, 12, 8, 10, 9, 13)> plot(sale5,new=T)> abline(lsfit(1:10,sale5))> abline(lsfit(1:10,sale5, intercept = FALSE), col= 4)> abline(h=6, v=8, col = "gray60")> text(8,6, "abline( h = 6 )", col = "gray60", adj = c(0, -.1))> abline(h = 4:8, v = 6:12, col = "lightgray", lty=3)> abline(a=1, b=2, col = 2)> text(5,11, "abline( 1, 2 )", col=2, adj=c(-.1,-.1))> segments(6,4,9,5,col="green")> text(6,5,"segments(6,4,9,5)")> lines(sale5,col="pink")

线段

box画出当前盒子的边界,polygon画多边形,polypath画路径,rect画距形。

> x <- c(1:9,8:1)> y <- c(1,2*(5:3),2,-1,17,9,8,2:9)> op <- par(mfcol=c(3,1))> for(xpd in c(FALSE,TRUE,NA)) {+   plot(1:10, main = paste("xpd =", xpd))+   box("figure", col = "pink", lwd=3)+   polygon(x,y, xpd=xpd, col="orange", lty=2, lwd=2, border="red")+ }> par(op)> plotPath <- function(x, y, col = "grey", rule = "winding") {+     plot.new()+     plot.window(range(x, na.rm = TRUE), range(y, na.rm = TRUE))+     polypath(x, y, col = col, rule = rule)+     if (!is.na(col))+         mtext(paste("Rule:", rule), side = 1, line = 0)+ }> > plotRules <- function(x, y, title) {+     plotPath(x, y)+     plotPath(x, y, rule = "evenodd")+     mtext(title, side = 3, line = 0)+     plotPath(x, y, col = NA)+ }> > op <- par(mfrow = c(5, 3), mar = c(2, 1, 1, 1))> > plotRules(c(.1, .1, .9, .9, NA, .2, .2, .8, .8),+           c(.1, .9, .9, .1, NA, .2, .8, .8, .2),+           "Nested rectangles, both clockwise")> plotRules(c(.1, .1, .9, .9, NA, .2, .8, .8, .2),+           c(.1, .9, .9, .1, NA, .2, .2, .8, .8),+           "Nested rectangles, outer clockwise, inner anti-clockwise")> plotRules(c(.1, .1, .4, .4, NA, .6, .9, .9, .6),+           c(.1, .4, .4, .1, NA, .6, .6, .9, .9),+           "Disjoint rectangles")> plotRules(c(.1, .1, .6, .6, NA, .4, .4, .9, .9),+           c(.1, .6, .6, .1, NA, .4, .9, .9, .4),+           "Overlapping rectangles, both clockwise")> plotRules(c(.1, .1, .6, .6, NA, .4, .9, .9, .4),+           c(.1, .6, .6, .1, NA, .4, .4, .9, .9),+           "Overlapping rectangles, one clockwise, other anti-clockwise")> > par(op)> require(grDevices)> ## set up the plot region:> op <- par(bg = "thistle")> plot(c(100, 250), c(300, 450), type = "n", xlab="", ylab="",+      main = "2 x 11 rectangles; 'rect(100+i,300+i,  150+i,380+i)'")> i <- 4*(0:10)> ## draw rectangles with bottom left (100, 300)+i> ## and top right (150, 380)+i> rect(100+i, 300+i, 150+i, 380+i, col=rainbow(11, start=.7,end=.1))> rect(240-i, 320+i, 250-i, 410+i, col=heat.colors(11), lwd=i/5)> ## Background alternating  ( transparent / "bg" ) :> j <- 10*(0:5)> rect(125+j, 360+j,   141+j, 405+j/2, col = c(NA,0),+      border = "gold", lwd = 2)> rect(125+j, 296+j/2, 141+j, 331+j/5, col = c(NA,"midnightblue"))> mtext("+  2 x 6 rect(*, col = c(NA,0)) and  col = c(NA,\"m..blue\"))")

多边形

路径

矩形

arrows用于画箭头，symbols用于画符号

> ## Note that  example(trees)  shows more sensible plots!> N <- nrow(trees)> with(trees, {+ ## Girth is diameter in inches+ symbols(Height, Volume, circles = Girth/24, inches = FALSE,+         main = "Trees' Girth") # xlab and ylab automatically+ ## Colours too:+ op <- palette(rainbow(N, end = 0.9))+ symbols(Height, Volume, circles = Girth/16, inches = FALSE, bg = 1:N,+         fg = "gray30", main = "symbols(*, circles = Girth/16, bg = 1:N)")+ palette(op)+ })

符号

画圆

> library(plotrix)> plot(1:5,seq(1,10,length=5),type="n",xlab="",ylab="",main="Test draw.circle")>  draw.circle(2,4,c(1,0.66,0.33),border="purple",+   col=c("#ff00ff","#ff77ff","#ffccff"),lty=1,lwd=1)>  draw.circle(2.5,8,0.6,border="red",lty=3,lwd=3)>  draw.circle(4,3,0.7,border="green",lty=1,lwd=1)>  draw.circle(3.5,7,0.8,border="blue",lty=2,lwd=2)

圆

散点图及趋势线

一维点图使用dotchart函数。

> # Dotplot: Grouped Sorted and Colored> # Sort by mpg, group and color by cylinder > x <- mtcars[order(mtcars$mpg),] # sort by mpg> x$cyl <- factor(x$cyl) # it must be a factor> x$color[x$cyl==4] <- "red"> x$color[x$cyl==6] <- "blue"> x$color[x$cyl==8] <- "darkgreen"> dotchart(x$mpg,labels=row.names(x),cex=.7,groups= x$cyl,+    main="Gas Milage for Car Models\ngrouped by cylinder",+    xlab="Miles Per Gallon", gcolor="black", color=x$color)

一维点图

二维散点图使用plot函数。直趋势线使用abline函数，拟合曲线在拟合后使用line函数绘制。

> attach(mtcars)> plot(wt, mpg, main="Scatterplot Example", +    xlab="Car Weight ", ylab="Miles Per Gallon ", pch=19)> # Add fit lines> abline(lm(mpg~wt), col="red") # regression line (y~x) > lines(lowess(wt,mpg), col="blue") # lowess line (x,y)

点阵图

曲线

曲线使用lines函数。其type参数可以使用”p”,”l”,”o”,”b,c”,”s,S”,”h”,”n”等。

> x <- c(1:5); y <- x # create some data> par(pch=22, col="blue") # plotting symbol and color > par(mfrow=c(2,4)) # all plots on one page > opts = c("p","l","o","b","c","s","S","h") > for(i in 1:length(opts)){ +   heading = paste("type=",opts[i]) +   plot(x, y, main=heading) +   lines(x, y, type=opts[i]) + }

曲线

柱状图

普通的柱状图使用barplot函数。其参数horiz=TRUE表示水平画图，beside=TRUE表示如果是多组数据的话，在并排画图，否则原位堆叠画图。

> par(mfrow=c(1,2))> counts <- table(mtcars$vs, mtcars$gear)> barplot(counts, main="Car Distribution by Gears and VS",+   xlab="Number of Gears", col=c("darkblue","red"),+   legend = rownames(counts),horiz=TRUE)> barplot(counts, main="Car Distribution by Gears and VS",+   xlab="Number of Gears", col=c("darkblue","red"),+   legend = rownames(counts), beside=TRUE)

bar图

柱状统计图使用hist函数。其breaks参数设置每组的范围。使用density函数可以拟合曲线。

> hist(mtcars$mpg, breaks=12)> dens<-density(mtcars$mpg)> lines(dens$x,dens$y*100,col="red")

柱状图

饼图

饼图使用pie函数。

> x<-table(mtcars$gear)> pie(x,label=paste("gear=",rownames(x),sep=""))

饼图

3维饼图使用plotrix库中的pie3D函数。

> x<-table(mtcars$gear)> pie3D(x,labels=paste("gear=",rownames(x),sep=""),explode=0.1)

3Ｄ饼图

箱线图

箱线图使用boxplot函数。boxplot中参数x为公式。R中的公式如何定义呢？最简单的 y ~ x 就是y是x的一次函数。好了，下面就是相关的符号代表的意思：

符号示例意义++x包括该变量–-x不包括该变量:x:z包括两变量的相互关系*x*z包括两变量，以及它们之间的相互关系/x/znesting: include z nested within x|x|z条件或分组：包括指定z的x^(u+v+w)^3include these variables and all interactions up to three waypolypoly(x,3)polynomial regression: orthogonal polynomialsErrorError(a/b)specify the error termII(x*z)as is: include a new variable consisting of these variables multiplied1-1截距：减去该截距

> boxplot(mpg~cyl,data=mtcars, main="Car Milage Data", +    xlab="Number of Cylinders", ylab="Miles Per Gallon")> boxplot(len~supp*dose, data=ToothGrowth, notch=TRUE, +   col=(c("gold","darkgreen")),+   main="Tooth Growth", xlab="Suppliment and Dose")

箱线图

如果我想在箱线图上叠加样品点，即所谓的蜂群图，如何做呢？

> source("http://bioconductor.org/biocLite.R")> biocLite(c("beeswarm","ggplot2"))> library(beeswarm)> library(ggplot2)> data(breast)> beeswarm <- beeswarm(time_survival ~ event_survival,+             data = breast, method = 'swarm',+             pwcol = ER)[, c(1, 2, 4, 6)]> colnames(beeswarm) <- c("x", "y", "ER", "event_survival")> > beeswarm.plot <- ggplot(beeswarm, aes(x, y)) ++   xlab("") ++   scale_y_continuous(expression("Follow-up time (months)"))> beeswarm.plot2 <- beeswarm.plot + geom_boxplot(aes(x, y,+   group = round(x)), outlier.shape = NA)> beeswarm.plot3 <- beeswarm.plot2 + geom_point(aes(colour = ER)) ++   scale_colour_manual(values = c("black", "red")) ++   scale_x_continuous(breaks = c(1:2),+ labels = c("Censored", "Metastasis"), expand = c(0, 0.5))> print(beeswarm.plot3)

蜂群图

> require(beeswarm)> data(breast)>  > beeswarm(time_survival ~ event_survival, data = breast,+         method = 'swarm',+         pch = 16, pwcol = as.numeric(ER),+         xlab = '', ylab = 'Follow-up time (months)',+         labels = c('Censored', 'Metastasis'))>  > boxplot(time_survival ~ event_survival, +         data = breast, add = T, +         names = c("",""), col="#0000ff22")

蜂群图

分枝树

> require(graphics)> opar<-par(mfrow=c(2,1),mar=c(4,3,0.5,0.5))> hc <- hclust(dist(USArrests), "ave")> plot(hc,main="")> plot(hc, hang = -1,main="")> par(opar)

分枝树

文氏图

> library(limma)> Y <- matrix(rnorm(100*6),100,6)> Y[1:10,3:4] <- Y[1:10,3:4]+3> Y[1:20,5:6] <- Y[1:20,5:6]+3> design <- cbind(1,c(0,0,1,1,0,0),c(0,0,0,0,1,1))> fit <- eBayes(lmFit(Y,design))> results <- decideTests(fit)> a <- vennCounts(results)> print(a)     x1 x2 x3 Counts[1,]  0  0  0     89[2,]  0  0  1     11[3,]  0  1  0      0[4,]  0  1  1      0[5,]  1  0  0      0[6,]  1  0  1      0[7,]  1  1  0      0[8,]  1  1  1      0attr(,"class")[1] "VennCounts"> vennDiagram(a)> vennDiagram(results,include=c("up","down"),counts.col=c("red","green"))

文氏图

七，综合应用

点阵图加两方向柱状图

> def.par <- par(no.readonly = TRUE) # save default, for resetting...> > x <- pmin(3, pmax(-3, rnorm(50)))> y <- pmin(3, pmax(-3, rnorm(50)))> xhist <- hist(x, breaks=seq(-3,3,0.5), plot=FALSE)> yhist <- hist(y, breaks=seq(-3,3,0.5), plot=FALSE)> top <- max(c(xhist$counts, yhist$counts))> xrange <- c(-3,3)> yrange <- c(-3,3)> nf <- layout(matrix(c(2,0,1,3),2,2,byrow=TRUE), c(3,1), c(1,3), TRUE)> #layout.show(nf)> > par(mar=c(3,3,1,1))> plot(x, y, xlim=xrange, ylim=yrange, xlab="", ylab="")> par(mar=c(0,3,1,1))> barplot(xhist$counts, axes=FALSE, ylim=c(0, top), space=0)> par(mar=c(3,0,1,1))> barplot(yhist$counts, axes=FALSE, xlim=c(0, top), space=0, horiz=TRUE)> > par(def.par)

点阵＋柱状

带误差的点线图

> # Clustered Error Bar for Groups of Cases. > # Example: Experimental Condition (Stereotype Threat Yes/No) x Gender (Male / Female)> # The following values would be calculated from data and are set fixed now for > # code reproduction> > means.females <- c(0.08306698, -0.83376319)> stderr.females <- c(0.13655378, 0.06973371)> > names(means.females) <- c("No","Yes")> names(stderr.females) <- c("No","Yes")> > means.males <- c(0.4942997, 0.2845608)> stderr.males <- c(0.07493673, 0.18479661)> > names(means.males) <- c("No","Yes")> names(stderr.males) <- c("No","Yes")> > # Error Bar Plot> > library (gplots)> > # Draw the error bar for female experiment participants:> plotCI(x = means.females, uiw = stderr.females, lty = 2, xaxt ="n", xlim = c(0.5,2.5), ylim = c(-1,1), gap = 0, ylab="Microworld Performance (Z Score)", xlab="Stereotype Threat", main = "Microworld performance over experimental conditions")> > # Add the males to the existing plot> plotCI(x = means.males, uiw = stderr.males, lty = 1, xaxt ="n", xlim = c(0.5,2.5), ylim = c(-1,1), gap = 0, add = TRUE)> > # Draw the x-axis (omitted above)> axis(side = 1, at = 1:2, labels = names(stderr.males), cex = 0.7)> > # Add legend for male and female participants> legend(2,1,legend=c("Male","Female"),lty=1:2)

误差图

误差绘图plotCI加强版是plotmeans,具体可使用?plotmeans来试用它。这里就不多讲了。

带误差的柱状图。还是使用gplots包

> library(gplots)> # Example with confidence intervals and grid> hh <- t(VADeaths)[, 5:1]> mybarcol <- "gray20"> ci.l <- hh * 0.85> ci.u <- hh * 1.15> mp <- barplot2(hh, beside = TRUE,+         col = c("lightblue", "mistyrose",+                 "lightcyan", "lavender"),+         legend = colnames(VADeaths), ylim = c(0, 100),+         main = "Death Rates in Virginia", font.main = 4,+         sub = "Faked 95 percent error bars", col.sub = mybarcol,+         cex.names = 1.5, plot.ci = TRUE, ci.l = ci.l, ci.u = ci.u,+         plot.grid = TRUE)> mtext(side = 1, at = colMeans(mp), line = 2,+       text = paste("Mean", formatC(colMeans(hh))), col = "red")> box()

带误差的柱状图

漂亮的箱线图

> require(gplots) #for smartlegend> > data(ToothGrowth)> boxplot(len ~ dose, data = ToothGrowth,+         boxwex = 0.25, at = 1:3 - 0.2,+         subset= supp == "VC", col="yellow",+         main="Guinea Pigs' Tooth Growth",+         xlab="Vitamin C dose mg",+         ylab="tooth length", ylim=c(0,35))> boxplot(len ~ dose, data = ToothGrowth, add = TRUE,+         boxwex = 0.25, at = 1:3 + 0.2,+         subset= supp == "OJ", col="orange")> > smartlegend(x="left",y="top", inset = 0,+             c("Ascorbic acid", "Orange juice"),+             fill = c("yellow", "orange"))

箱线图

基因芯片热图，具体参考http://www2.warwick.ac.uk/fac/sci/moac/students/peter_cock/r/heatmap/

> library("ALL")> data("ALL")> eset <- ALL[, ALL$mol.biol %in% c("BCR/ABL", "ALL1/AF4")]> library("limma")> f <- factor(as.character(eset$mol.biol))> design <- model.matrix(~f)> fit <- eBayes(lmFit(eset,design))> selected  <- p.adjust(fit$p.value[, 2]) <0.005> esetSel <- eset [selected, ]> color.map <- function(mol.biol) { if (mol.biol=="ALL1/AF4") "#FF0000" else "#0000FF" }> patientcolors <- unlist(lapply(esetSel$mol.bio, color.map))> library("gplots")> heatmap.2(exprs(esetSel), col=redgreen(75), scale="row", ColSideColors=patientcolors,+            key=TRUE, symkey=FALSE, density.info="none", trace="none", cexRow=0.5)

点柱图(dothistogram)

英文名称中文名称bar条形图line线图area面积图pie饼图high-low高低图pareto帕累托图control控制图boxplot箱线图error bar误差条图scatter散点图P-PP-P正态概率图Q-QQ-Q正态概率图sequence序列图ROC CurveROC分类效果曲线图Time Series时间序列图

好了，言归正传。那么什么又是点柱图(dot histogram)呢？之前我又称之为蜂群图(beeswarm)。还有称之为抖点图(jitter plots)。总之无论如何，在糗世界里我都称之为点柱图吧。

我们先看点柱图效果：

点柱图

以下是代码

> require(beeswarm)> data(breast)> head(breast)            ER      ESR1     ERBB2 time_survival event_survival100.CEL.gz neg  8.372133 13.085894            39              1103.CEL.gz pos 10.559356  9.491683            97              0104.CEL.gz pos 12.299905  9.599574            11              1105.CEL.gz pos 10.776632  9.681747            99              0106.CEL.gz pos 10.505124  9.436763            40              1107.CEL.gz neg 10.377741  8.695576            94              0> require(plotrix)> cluster<-cluster.overplot(breast$event_survival, breast$time_survival)> png("dothist.png",width=1000,height=1000)> opar<-par(mfrow=c(3,3))> plot (breast$event_survival, breast$time_survival, main="Multiple points on coordinate",col=as.numeric(breast$ER),xaxt="n",xlim=c(-1,2))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> plot(jitter(breast$event_survival), breast$time_survival, main="Using Jitter on x-axis",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> plot(jitter(breast$event_survival), jitter(breast$time_survival), main="Using Jitter on x and y-axis",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> sunflowerplot(breast$event_survival, breast$time_survival, main="Using Sunflowers",xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> plot(cluster, main="Using cluster.overplot",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> count.overplot(jitter(breast$event_survival), jitter(breast$time_survival), main="Using cout.overplot",col=as.numeric(breast$ER),xaxt="n")> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> sizeplot(breast$event_survival, breast$time_survival, main="Using sizeplot",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> beeswarm(time_survival ~ event_survival, data = breast,+          method = 'swarm',+          pch = 16, pwcol = as.numeric(ER),+          xlab = '', ylab = 'Follow-up time (months)',+          labels = c('Censored', 'Metastasis'))> dev.off()quartz     2> par(opar)

以下是解释

在很多情况下，我们画散点图的时候，有许多点拥有相同的横坐标，如果我们简单的使用plot(x,y)的方式，会显得这些点拥挤在一起，象图中左上角一样，非常的不舒服。我们需要把这些点分散开。

最基本的思路是，把横坐标抖散(jitter)，使本来都拥有相同坐标的点的横坐标稍有不同。jitter是基类函数{base}，无需调用任何包。

> plot(jitter(breast$event_survival), breast$time_survival, main="Using Jitter on x-axis",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))> plot(jitter(breast$event_survival), jitter(breast$time_survival), main="Using Jitter on x and y-axis",col=as.numeric(breast$ER),xaxt="n",xlim=c(-0.5,1.5))> axis(1,at=c(0,1),labels=c("Censored","Metastasis"))

我们比较图中上边靠右的两个图，我们发现，如果只抖散x坐标的话，还是有些点会粘在一起，所以同时抖散y坐标会好一些。我们可以使用factor参数来控制jitter抖散的强度。

> plot(rep(c(1,5,10),each=5),c(jitter(rep(100,5),factor=1),jitter(rep(100,5),factor=5),jitter(rep(100,5),factor=10)),col=c("red","blue","green","gray","black"),xlim=c(-2,13),xlab="",ylab="y",xaxt="n",main="jitter(rep(100,5)) with different factor")> axis(1,at=c(1,5,10),labels=c(paste("factor=",c(1,5,10),sep="")))

不同强度的jitter

在graphics包中提供了一个sunflowerplot的函数。它的目的是用花瓣数目多少来显示落在同一坐标上的点的数目。但是从中左图看来，点多的时候效果并非总是那么好。

在plotrix包中提供了一些有意思的函数来解决点挤在一起的这个问题，它们分别是cluster.overplot, count.overplot, sizeplot。这三个函数的效果如图中及下靠左的两个。cluster.overplot的方法类似抖散，count overplot的方法是使用数字来显示落在同一坐标上的点的数目，sizeplot的方法是使用不同大小是点来显示落在同一坐标上的点的数目。从效果来看，点多的时候效果也并非理想。

而上一次提到过的蜂群图似乎是解决这一问题的最佳方案。

我们得出结论，在点数不同的情况下，使用plotrix包及sunflowerplot是不错的。但点数较多的情况下，还是使用jitter和beeswarm较为稳妥。

我们也可以使用ggplot2包中的geom来绘制点柱图

> require(beeswarm)> data(breast)> library(ggplot2)> p<-ggplot(breast, aes(event_survival,time_survival))> print(p+geom_jitter(aes(color=ER))+scale_colour_manual(value = c("black", "red"))+scale_x_continuous(breaks = c(0:1),labels = c("Censored", "Metastasis")))

ggplot点柱图

还有一种图，名称为Engelmann-Hecker-Plot，由plotrix的ehplot来实现。

> data(iris);library(plotrix)> ehplot(iris$Sepal.Length, iris$Species,+ intervals=20, cex=1.8, pch=20, main="pch=20")> ehplot(iris$Sepal.Width, iris$Species,+ intervals=20, box=TRUE, median=FALSE, main="box=TRUE")> ehplot(iris$Petal.Length, iris$Species,+ pch=17, col="red", log=TRUE, main="pch=17")> ehplot(iris$Petal.Length, iris$Species,+ offset=0.06, pch=as.numeric(iris$Species), main="pch=as.numeric(iris$Species)")> rnd <- sample(150)> plen <- iris$Petal.Length[rnd]> pwid <- abs(rnorm(150, 1.2))> spec <- iris$Species[rnd]> ehplot(plen, spec, pch=19, cex=pwid,+ col=rainbow(3, alpha=0.6)[as.numeric(spec)], main="cex and col changes")

坐标中断(axis breaks)

R当中的坐标中断一般都使用plotrix库中的axis.break(), gap.plot(), gap.barplot(), gap.boxplot()等几个函数来实现，例：

gap plot

> library(plotrix)> opar<-par(mfrow=c(3,2))> plot(sample(5:7,20,replace=T),main="Axis break test",ylim=c(2,8))> axis.break(axis=2,breakpos=2.5,style="gap")> axis.break(axis=2,breakpos=3.5,style="slash")> axis.break(axis=2,breakpos=4.5,style="zigzag")> twogrp<-c(rnorm(5)+4,rnorm(5)+20,rnorm(5)+5,rnorm(5)+22)> gap.plot(twogrp,gap=c(8,16,25,35),+  xlab="X values",ylab="Y values",xlim=c(1,30),ylim=c(0,25),+  main="Test two gap plot with the lot",xtics=seq(0,30,by=5),+  ytics=c(4,6,18,20,22,38,40,42),+  lty=c(rep(1,10),rep(2,10)),+  pch=c(rep(2,10),rep(3,10)),+  col=c(rep(2,10),rep(3,10)),+  type="b")> gap.plot(21:30,rnorm(10)+40,gap=c(8,16,25,35),add=TRUE,+   lty=rep(3,10),col=rep(4,10),type="l")> gap.barplot(twogrp,gap=c(8,16),xlab="Index",ytics=c(3,6,17,20),+   ylab="Group values",main="Barplot with gap")> gap.barplot(twogrp,gap=c(8,16),xlab="Index",ytics=c(3,6,17,20),+   ylab="Group values",horiz=TRUE,main="Horizontal barplot with gap")> twovec<-list(vec1=c(rnorm(30),-6),vec2=c(sample(1:10,40,TRUE),20))>  gap.boxplot(twovec,gap=list(top=c(12,18),bottom=c(-5,-3)),+  main="Show outliers separately")> gap.boxplot(twovec,gap=list(top=c(12,18),bottom=c(-5,-3)),range=0,+  main="Include outliers in whiskers")> par(opar)

从图像效果上来看，这样的坐标中断只能说实现了坐标中断，但效果上是非常一般的。甚至远不如excel, openoffice当中出图效果好。为此，我们需要对plotrix库中的gap.plot做出修改，以达到满意的效果。

最简单的修改办法就是在使用了gap.plot, gap.barplot, gap.boxplot之后重新使用axis.break来修改中断类型，使得看上去美一点。

> axis.break(2,from,breakcol="snow",style="gap")> axis.break(2,from*(1+0.02),breakcol="black",style="slash")> axis.break(4,from*(1+0.02),breakcol="black",style="slash")> axis(2,at=from)

使用上面的办法可以绘制出双反斜线中断，并可以视实际情况加油断点起止位置。

改进的gap.plot

> library(plotrix)> x<-c(1:5,6.9,7)> y<-2^x> from<-33> to<-110> opar<-par(mfrow=c(2,2))> plot(x,y,type="b",main="normal plot")> gap.plot(x,y,gap=c(from,to),type="b",main="gap plot")> axis.break(2,from,breakcol="snow",style="gap")> axis.break(2,from*(1+0.02),breakcol="black",style="slash")> axis.break(4,from*(1+0.02),breakcol="black",style="slash")> axis(2,at=from)> gap.barplot(y,gap=c(from,to),col=as.numeric(x),main="barplot with gap")> axis.break(2,from,breakcol="snow",style="gap")> axis.break(2,from*(1+0.02),breakcol="black",style="slash")> axis.break(4,from*(1+0.02),breakcol="black",style="slash")> axis(2,at=from)> gap.barplot(y,gap=c(from,to),col=as.numeric(x),horiz=T,main="Horizontal barplot with gap")> axis.break(1,from,breakcol="snow",style="gap")> axis.break(1,from*(1+0.02),breakcol="black",style="slash")> axis.break(3,from*(1+0.02),breakcol="black",style="slash")> axis(1,at=from) > par(opar)

来源：http://blog.qiubio.com:8080/rbioconductor-2

0 0