spia(de=NULL,all=NULL,organism="hsa",data.dir=NULL,pathids=NULL,nB=2000,plots=FALSE,verbose=TRUE,beta=NULL,combine="fisher")de argument.c("activation","compound","binding/association","expression","inhibition","activation_phosphorylation","phosphorylation",
"indirect","inhibition_phosphorylation","dephosphorylation_inhibition","dissociation","dephosphorylation","activation_dephosphorylation",
"state","activation_indirect","inhibition_ubiquination","ubiquination","expression_indirect","indirect_inhibition","repression",
"binding/association_phosphorylation","dissociation_phosphorylation","indirect_phosphorylation")If set to null, beta will be by default chosen as: c(1,0,0,1,-1,1,0,0,-1,-1,0,0,1,0,1,-1,0,1,-1,-1,0,0,0).
"fisher" it will use Fisher's method. If set to "norminv" it will use the normal inversion method.pSize is the number of genes on the pathway;
NDE is the number of DE genes per pathway; tA is the observed total preturbation
accumulation in the pathway; pNDE is the probability to observe at least NDE genes on
the pathway using a hypergeometric model;
pPERT is the probability to observe a total accumulation more extreme than tA only by
chance;
pG is the p-value obtained by combining pNDE and pPERT;
pGFdr and pGFWER are the
False Discovery Rate and respectively Bonferroni adjusted global p-values; and the Status gives the direction
in which the pathway is perturbed (activated or inhibited).
KEGGLINK gives a web link to the KEGG website that displays the pathway image with the differentially expressed genes
highlighted in red.
Purvesh Khatri, Sorin Draghici, Adi L. Tarca, Sonia S. Hassan, Roberto Romero. A system biology approach for the steady-state analysis of gene signaling networks. Progress in Pattern Recognition, Image Analysis and Applications, Lecture Notes in Computer Science. 4756:32-41, November 2007.
Draghici, S., Khatri, P., Tarca, A.L., Amin, K., Done, A., Voichita, C., Georgescu, C., Romero, R.: A systems biology approach for pathway level analysis. Genome Research, 17, 2007.
plotP# Example using a colorectal cancer dataset obtained using Affymetrix geneChip technology (GEE GSE4107).
# Suppose that proper preprocessing was performed and a two group moderated t-test was applied. The topTable
# result from limma package for this data set is called "top".
#The following lines will annotate each probeset to an entrez ID identifier, will keep the most significant probeset for each
#gene ID and retain those with FDR<0.05 as differentially expressed.
#You can run these lines if hgu133plus2.db package is available
#data(colorectalcancer)
#x <- hgu133plus2ENTREZID
#top$ENTREZ<-unlist(as.list(x[top$ID]))
#top<-top[!is.na(top$ENTREZ),]
#top<-top[!duplicated(top$ENTREZ),]
#tg1<-top[top$adj.P.Val<0.1,]
#DE_Colorectal=tg1$logFC
#names(DE_Colorectal)<-as.vector(tg1$ENTREZ)
#ALL_Colorectal=top$ENTREZ
data(colorectalcancer)
# pathway analysis using SPIA; # use nB=2000 or higher for more accurate results
#uses older version of KEGG signalimng pathways graphs
res<-spia(de=DE_Colorectal, all=ALL_Colorectal, organism="hsa",beta=NULL,nB=2000,plots=FALSE, verbose=TRUE,combine="fisher")
res
# Create the evidence plot
plotP(res)
#now combine pNDE and pPERT using the normal inversion method without running spia function again
res$pG=combfunc(res$pNDE,res$pPERT,combine="norminv")
res$pGFdr=p.adjust(res$pG,"fdr")
res$pGFWER=p.adjust(res$pG,"bonferroni")
plotP(res,threshold=0.05)
#highlight the colorectal cancer pathway in green
points(I(-log(pPERT))~I(-log(pNDE)),data=res[res$ID=="05210",],col="green",pch=19,cex=1.5)
#run SPIA using pathways data generated from (up-to-date) xml files that you can obtain from
#KEGG ftp or by downloading them from each pathway's web page:
# e.g. go to http://www.genome.jp/kegg/pathway/hsa/hsa04010.html and click on DOwnload KGML
#to get the xml file for pathway 4010
makeSPIAdata(kgml.path=system.file("extdata/keggxml/hsa",package="SPIA"),organism="hsa",out.path="./")
res<-spia(de=DE_Colorectal, all=ALL_Colorectal, organism="hsa",data.dir="./")
res
Run the code above in your browser using DataLab