R包是否应该每次使用都联网？以及生信必备统计学实例推荐

  ###   over-representation test
  kk.up <- enrichKEGG(gene         = gene_up,
                      organism     = 'hsa',
                      universe     = gene_all,
                      pvalueCutoff = 0.9,
                      qvalueCutoff =0.9)
  head(kk.up)[,1:6]

library(ggplot2) 
library(KEGG.db)
library(org.Hs.eg.db)
load('for_annotation.Rdata')
## KEGG pathway analysis
library(KEGG.db)
tmp=toTable(org.Hs.egPATH)

minS=10;
maxS=500;
GeneID2Path<<- tapply(tmp[,2],as.factor(tmp[,1]),function(x) x)
Path2GeneID<<- tapply(tmp[,1],as.factor(tmp[,2]),function(x) x)
# 超几何分布检验是统计学上一种离散概率分布。它描述了由有限个物件中抽出n个物件，成功抽出指定种类的物件的个数（不归还）
gened=intersect(names(GeneID2Path),gene_up)  ## all genes been choosed
geneb=intersect(names(GeneID2Path),gene_all) ## all genes as backgroud 
N=length(geneb);N ## number of all the genes
M=length(gened);M ## number of all the choosed genes

kegg_r <- do.call(rbind,lapply(names(Path2GeneID),function(thisP){
  # thisP=names(Path2GeneID)[1];thisP
  n=length(intersect(geneb,Path2GeneID[[thisP]]));n
  if(n>maxS) return(NULL);
  if(n<minS) return(NULL);
  (  exp_count=n*M/N)
  k=length(intersect(gened,Path2GeneID[[thisP]]));k
  OddsRatio=k/exp_count 
  p=phyper(k-1,M, N-M, n, lower.tail=F)
  #print(paste(i,p,OddsRatio,exp_count,k,M,sep="    "))
  return(c(thisP,p,OddsRatio,exp_count,k,n,M,N))
}))
colnames(kegg_r)=c('path_id','p','OddsRatio','exp_count','k','n','M','N')
kegg_r=merge(kegg_r,toTable(KEGGPATHID2NAME),by='path_id')
kegg_r=kegg_r[order(kegg_r$p),]