MPINet (version 1.0)

identifypathway: A novel pathway identification approach based on metabolite set

Description

Identify pathways via global weighted human metabolite network, which considering both the global non-equivalence of metabolites in pathway and the bias existing in metabonomic experiment technology.

Usage

identifypathway(componentList,PSS,pathType="KEGG",method="MPINet",weightnum=6, backgroundcid=Getenvir("getBackground"),annlim=1,bglim=6, order="pvalue",decreasing=FALSE)

Arguments

componentList

A character vector of interesting metabolites, for each element is a pubchem CID.

PSS

A data frame, which is obtained by the function getPSS.

pathType

A character string vector specifying the pathway source,must be some of the elements in the vector c("KEGG", "consensusPath", "PharmGKB", "SMPDB", "Wikipathways", "PID", "Reactome", "INOH", "BioCarta", "HumanCyc", "EHMN"). The default is KEGG.

method

A character string specifying the pathway identification method, must be one of "MPINet"(default) and "Hyper".

weightnum

A value. The power of the relative weight of pathway. The default value is 6.

backgroundcid

A character vector of the background metabolites, which is used to identify the statistically significant pathways.

annlim

An integer. Only use pathways annotated at least this number of metabolites. The default value is 1.

bglim

An integer. Only use pathways containing at least this number of metabolites. The default value is 6.

order

A character string. Should be one of "pvalue" and "fdr".

decreasing

A logical. Should the sort order be increasing or decreasing?

Value

A list. Each element of the list is another list. (i)If the argument method is "MPINet", it includes the following elements: 'pathwayName', 'annComponentList', 'annComponentNumber', 'annBgComponentList', 'annBgNumber', 'componentNumber', 'bgNumber', 'pvalue', 'fdr', 'InWeight', 'weight', 'anncompinNetworkNum', 'anncompinNetworkList', 'riskcompinNetworkNum', 'riskcompinNetworkList'. They correspond to pathway name, the submitted metabolites annotated to a pathway, numbers of submitted metabolites annotated to a pathway, the background metabolites annotated to a pathway, numbers of background metabolites annotated to a pathway, numbers of submitted metabolites, numbers of background metabolitess, p-value of the Wallenius' noncentral hypergeometric test, Benjamini-Hochberg fdr values, the mean score value of metabolites in pathway, the final weight of pathway, numbers of the submitted metabolites annotated to a pathway and in the global human metabolite network, the submitted metabolites annotated to a pathway and in the global human metabolite network, numbers of submitted metabolites in the global human metabolite network, submitted metabolites in the global human metabolite network. When the argument pathType is "KEGG", the 'pathwayId' element is also included, which is the pathway identifier in KEGG. When the argument pathType is not "KEGG", the 'pathsource' element is also included, which stands for the source of pathway. (ii)If the argument method is "Hyper", it includes the same elements as (i), but not includes the following elements: 'InWeight', 'weight', 'anncompinNetworkNum', 'anncompinNetworkList', 'riskcompinNetworkNum', 'riskcompinNetworkList'. To save the results, the list can be converted to the data.frame by the function printGraph.Note that componentList submitted by users must be a 'character' vector.

Details

The function can annotate a set of metabolites to pathways and identify the statistically significantly enriched pathways. The argument method should be one of "MPINet" and "Hyper".When the "MPINet" is specified, MPINet method which is considers both the global non-equivalence of metabolites in pathway and bias existing in metabonomic experiment technology is used. When the "Hyper" method is selected, the Hypergeometric test is used. If users don't set the values of the argument backgroundcid, the human background metabolites will be obtained from our default data set which contains 4994 metabolites and selected from five databases including MSEA, HMDB, SMPDB, KEGG and Reactome. Note that the argument weightnum can be assigned according to the bias level, which can be evaluated through the getPSS function by set the plot argument as "TRUE". If the plot line is more closer to right up, the weightnum should be assigned a higher value.

References

Subramanian, A., Tamayo, P., Mootha, V.K., Mukherjee, S., Ebert, B.L., Gillette, M.A., Paulovich, A., Pomeroy, S.L., Golub, T.R., Lander, E.S. et al. (2005) Gene set enrichment analysis: a knowledge-based approach for interpreting genome-wide expression profiles. Proc Natl Acad Sci U S A, 102, 15545-15550.

Li, C., Li, X., Miao, Y., Wang, Q., Jiang, W., Xu, C., Li, J., Han, J., Zhang, F., Gong, B. et al. (2009) SubpathwayMiner: a software package for flexible identification of pathways. Nucleic Acids Res, 37, e131.

Examples

## Not run: 
# 
# #####identify pathways related with metastatic prostate cancer###########
# #example 1
# #get example data
# #### get the metastatic prostate cancer interesting metabolite data set
# risk<-GetExampleData(dataset="prostate") 
# #### integrate the global non-equivalence of metabolites and the character of 
# #### differential metabolites by the monotonic spline model
# pss<-getPSS(risk)         
# 
# #identify pathways
# anncpdpre<-identifypathway(risk,pss,pathType="KEGG",method="MPINet",annlim=1,bglim=6)
# #convert ann to data.frame
# result<-printGraph(anncpdpre,pathType="KEGG",method="MPINet")
# #print part of the results to screen
# head(result)
# 
# ##write the results to tab delimited file. 
# write.table(result,file="result.txt",row.names=FALSE,sep="\t")
# 
# result1<-printGraph(anncpdpre,pathType="KEGG",method="MPINet",detail=TRUE)
# ##write the results to tab delimited file. 
# write.table(result1,file="result1.txt",row.names=FALSE,sep="\t")
# 
# 
# #example 2
# #get example data from file
# risk<-read.table(paste(system.file(package="MPINet"),"/localdata/prostate.txt",sep=""),
# header=F,sep="\t","\"")
# 
# ####convert the data to a character vector
# risk<-as.character(risk[[1]])
# 
# pss<-getPSS(risk)  
# 
# #identify pathways
# anncpdpre<-identifypathway(risk,pss,pathType="KEGG",method="MPINet",annlim=1,bglim=6)
# #convert ann to data.frame
# result<-printGraph(anncpdpre,pathType="KEGG",method="MPINet")
# #print part of the results to screen
# head(result)
# 
# #example 3
# #get example data
# #### get the metastatic prostate cancer interesting metabolite data set
# risk<-GetExampleData(dataset="prostate") 
# pss<-getPSS(risk)  
# 
# #identify dysregulated Reactome and KEGG pathways
# anncpdpre<-identifypathway(risk,pss,pathType=c("KEGG","Reactome"),
# method="MPINet",annlim=1,bglim=6)
# #convert ann to data.frame
# result<-printGraph(anncpdpre,pathType=c("KEGG","Reactome"),method="MPINet")
# #print part of the results to screen
# head(result)
# 
# 
# 
# ## End(Not run)