R语言绘图相关性分析及绘图展示.docx

相关性分析
gaom
在我们平时分析的时候，经常会遇到样品间的相关性检验分析，并以此判断对我们后续分析的影响。今天主要跟大家讨论一下简单的相关性分析以及结果展示。利用的测试数据还是之前我们在geo数据库中随便找的一份表达谱数据。首先还是导入数据，进行简单分析，获取相关数值。
rm(list=ls())#先把我们的R清空一下data<-read.table(file="C:\\Users\\gaom\\Desktop\\R语言绘图\\相关性分析\\test_data.txt",header=T,row.names=1,sep="\t")#读取数据cor(data,method="pearson")#方法可选pearson、kendall、spearman。
##T01T02T03T04T05T06T07##T011.00000000.96268780.98205870.97756370.96728880.96641560.9752635##T020.96268781.00000000.98717930.97399350.97791550.97941410.9786400##T030.98205870.98717931.00000000.98235760.98196840.98080630.9833352##T040.97756370.97399350.98235761.00000000.99156930.98909070.9815730##T050.96728880.97791550.98196840.99156931.00000000.99430360.9805366##T060.96641560.97941410.98080630.98909070.99430361.00000000.9798487##T070.97526350.97864000.98333520.98157300.98053660.97984871.0000000##T080.97148010.97913690.98164820.98146550.98044640.97965990.9938647##T090.97464750.98023580.98451140.98148570.98007460.97949350.9947428##T100.96364980.97179020.97576520.97599360.97710730.97559220.9850377##T110.97397320.96770720.97755760.97969790.97787220.97602310.9876567##T120.96131860.96851090.97323890.97391450.97572720.97372340.9855969##T08T09T10T11T12##T010.97148010.97464750.96364980.97397320.9613186##T020.97913690.98023580.97179020.96770720.9685109##T030.98164820.98451140.97576520.97755760.9732389##T040.98146550.98148570.97599360.97969790.9739145##T050.98044640.98007460.97710730.97787220.9757272##T060.97965990.97949350.97559220.97602310.9737234##T070.99386470.99474280.98503770.98765670.9855969##T081.00000000.99422970.98581700.98490120.9837345##T090.99422971.00000000.98491670.98508920.9839299##T100.98581700.98491671.00000000.98676870.9878700##T110.98490120.98508920.98676871.00000000.9919710##T120.98373450.98392990.98787000.99197101.0000000
获得每个样品之间的相关系数，下面让我们把这些结果可视化。先利用基本函数plot展示下我们样品的结果。
plot(data[,1:6],pch=19,col=rgb(0,0,100,50,maxColorValue=255))#我们截取前面6个样品看看他们的相关性

从上面的结果来看T01、T02、T03之间的相关性要稍微差些，而T04、T05、T06之间的相关性要好的多。下面我们根据cor函数的结果做成类似热图的形式。