R常用统计 - 相关关系分析

数据格式

每行分别为表型和基因表达情况对应标量，每列分别为样品名的矩阵。假定前9列为phenotype，从第10行起为gene_id,编写简单for循环如下：

script

install.package("lessR")
library(lessR)
data = read.table("c:/Users/****/Desktop/yourfile.txt",
                  header = T,row.names = 1,na.strings = T,sep = "\t") #input your file
results = NULL
for (i in 1:9) {#
   for (j in 10:nrow(data)){
    if (median(as.numeric(data[j,]))>=1e-5) {
    x = data[i,]
    y = data[j,]
    tem_res = Correlation(x,
                          y, 
                          meth = "spearman",
                          adjt = "fdr",
                          heat_map = F)
  term_result = data.frame(row.names(data)[i],row.names(data)[j],tem_res$r,tem_res$pvalue)
  results = rbind(results,term_result)  
     }
  }
}
colnames(results) = c("phenotype","gene_id","r_value","p_value")
write.csv(results,"correlation_results.csv",row.names = F)