| Title: | Automate the Mapping Between a List of Genes and Gene Ontology Categories |
|---|---|
| Description: | In gene-expression microarray studies, for example, one generally obtains a list of dozens or hundreds of genes that differ in expression between samples and then asks 'What does all of this mean biologically?' Alternatively, gene lists can be derived conceptually in addition to experimentally. For instance, one might want to analyze a group of genes known as housekeeping genes. The work of the Gene Ontology (GO) Consortium <geneontology.org> provides a way to address that question. GO organizes genes into hierarchical categories based on biological process, molecular function and subcellular localization. The role of 'GoMiner' is to automate the mapping between a list of genes and GO, and to provide a statistical summary of the results as well as a visualization. |
| Authors: | Barry Zeeberg [aut, cre] |
| Maintainer: | Barry Zeeberg <barryz2013@gmail.com> |
| License: | GPL (>= 2) |
| Version: | 1.0 |
| Built: | 2025-03-25 11:19:06 UTC |
| Source: | CRAN |
compute the false discovery rate (FDR) of the hypergeometric p values of genes mapping to gene ontology (GO) categories
FDR(sampleList, GOGOA3, nrand, ONT)FDR(sampleList, GOGOA3, nrand, ONT)
sampleList |
character vector of user-supplied genes of interest |
GOGOA3 |
return value of subsetGOGOA() |
nrand |
integer number of randomizations |
ONT |
c("molecular_function","cellular_component","biological_process") |
returns a list with FDR information
#load("data/GOGOA3small.RData") sampleList<-randSubsetGeneList(GOGOA3small$genes[["biological_process"]],10) fdr<-FDR(sampleList,GOGOA3small,nrand=100,"biological_process")#load("data/GOGOA3small.RData") sampleList<-randSubsetGeneList(GOGOA3small$genes[["biological_process"]],10) fdr<-FDR(sampleList,GOGOA3small,nrand=100,"biological_process")
compute the gene enrichment in a GO category
GOenrich3(tableSample3, tablePop3)GOenrich3(tableSample3, tablePop3)
tableSample3 |
sample return value of GOtable3() |
tablePop3 |
population return value of GOtable3() |
returns a matrix with columns c("SAMPLE","POP","ENRICHMENT")
#load("data/x_tableSample3.RData") #load("data/x_tablePop3.RData") m<-GOenrich3(x_tableSample3,x_tablePop3)#load("data/x_tableSample3.RData") #load("data/x_tablePop3.RData") m<-GOenrich3(x_tableSample3,x_tablePop3)
generate a matrix to be used as input to a heat map
GOheatmap(sampleList, x, thresh, fdrThresh = 0.105)GOheatmap(sampleList, x, thresh, fdrThresh = 0.105)
sampleList |
character list of gene names |
x |
DB component of return value of GOtable3() |
thresh |
output of GOthresh() |
fdrThresh |
numeric value of FDR acceptance threshold |
returns a matrix to be used as input to a heat map
## Not run: # GOGOA3.RData is too large to include in the R package # you can generate it using the package 'minimalistGODB' # or you can retrieve it from https://github.com/barryzee/GO #load("~/GODB_RDATA/GOGOA3.RData") ONT<-"biological_process" DB<-GOGOA3$ontologies[[ONT]] #load("data/cluster52.RData") sampleList<-cluster52 #load("data/x_thresh.RData") heatmap<-GOheatmap(sampleList,DB,x_thresh) ## End(Not run)## Not run: # GOGOA3.RData is too large to include in the R package # you can generate it using the package 'minimalistGODB' # or you can retrieve it from https://github.com/barryzee/GO #load("~/GODB_RDATA/GOGOA3.RData") ONT<-"biological_process" DB<-GOGOA3$ontologies[[ONT]] #load("data/cluster52.RData") sampleList<-cluster52 #load("data/x_thresh.RData") heatmap<-GOheatmap(sampleList,DB,x_thresh) ## End(Not run)
compute the hypergeometric p value for gene enrichment in a GO category
GOhypergeometric3(tableSample3, tablePop3)GOhypergeometric3(tableSample3, tablePop3)
tableSample3 |
sample return value of GOtable3() |
tablePop3 |
population return value of GOtable3() |
returns a matrix with columns c("x","m","n","k","p")
#load("data/x_tableSample3.RData") #load("data/x_tablePop3.RData") hyper<-GOhypergeometric3(x_tableSample3,x_tablePop3)#load("data/x_tableSample3.RData") #load("data/x_tablePop3.RData") hyper<-GOhypergeometric3(x_tableSample3,x_tablePop3)
driver to generate heatmap
GoMiner( title = NULL, dir, sampleList, GOGOA3, ONT, enrichThresh = 2, countThresh = 5, fdrThresh = 0.1, nrand = 100 )GoMiner( title = NULL, dir, sampleList, GOGOA3, ONT, enrichThresh = 2, countThresh = 5, fdrThresh = 0.1, nrand = 100 )
title |
character string descriptive title |
dir |
character string full pathname to the directory acting result repository |
sampleList |
character list of gene names |
GOGOA3 |
return value of subsetGOGOA() |
ONT |
character string c("molecular_function", "cellular_component", "biological_process") |
enrichThresh |
numerical acceptance threshold for enrichment |
countThresh |
numerical acceptance threshold for gene count |
fdrThresh |
numerical acceptance threshold for fdr |
nrand |
numeric number of randomizations to compute FDR |
returns a matrix suitable to generate a heatmap
## Not run: # GOGOA3.RData is too large to include in the R package # you can generate it using the package 'minimalistGODB' # or you can retrieve it from https://github.com/barryzee/GO load("~/GODB_RDATA/GOGOA3.RData") load("data/cluster52.RData") l<-GoMiner("Cluster52",tempdir(),cluster52, GOGOA3,ONT="biological_process",enrichThresh=2, countThresh=5,fdrThresh=0.10,nrand=10) ## End(Not run)## Not run: # GOGOA3.RData is too large to include in the R package # you can generate it using the package 'minimalistGODB' # or you can retrieve it from https://github.com/barryzee/GO load("~/GODB_RDATA/GOGOA3.RData") load("data/cluster52.RData") l<-GoMiner("Cluster52",tempdir(),cluster52, GOGOA3,ONT="biological_process",enrichThresh=2, countThresh=5,fdrThresh=0.10,nrand=10) ## End(Not run)
tabulate number of geneList mappings to GO categories
GOtable3(hgncList, DB)GOtable3(hgncList, DB)
hgncList |
character list of gene names |
DB |
selected ontology branch of return value of subsetGOGOA |
returns a list whose components are c("DB","table","ngenes") where 'DB' is the GO DB subsetted to the desired ONTOLOGY, and 'table' is tabulation of number of occurrences of each GO category name within the desired ONTOLOGY, and ngenes is the total number of hgncList genes mapping to GOGOA
#load("data/GOGOA3small.RData") DB<-GOGOA3small$ontologies[["biological_process"]] # housekeeping genes downloaded from https://housekeeping.unicamp.br/?download #load("data/Housekeeping_Genes.RData") hgncList<-Housekeeping_Genes[,"Gene.name"] x<-GOtable3(hgncList,DB)#load("data/GOGOA3small.RData") DB<-GOGOA3small$ontologies[["biological_process"]] # housekeeping genes downloaded from https://housekeeping.unicamp.br/?download #load("data/Housekeeping_Genes.RData") hgncList<-Housekeeping_Genes[,"Gene.name"] x<-GOtable3(hgncList,DB)
retrieve lines of m that meet both enrichThresh and countThresh
GOthresh(m, sampleFDR, enrichThresh, countThresh, fdrThresh)GOthresh(m, sampleFDR, enrichThresh, countThresh, fdrThresh)
m |
return value of GOenrich3() |
sampleFDR |
component of return value of RCPD() |
enrichThresh |
numerical acceptance threshold for enrichment |
countThresh |
numerical acceptance threshold for gene count |
fdrThresh |
numerical acceptance threshold for fdr |
returns a subset of matrix (m joined with fdr$sampleFDR) with entries meeting all thresholds
#load("data/x_m.RData") #load("data/x_fdr.RData") thresh<-GOthresh(x_m,x_fdr$sampleFDR,enrichThresh=2,countThresh=2,fdrThresh=0.100)#load("data/x_m.RData") #load("data/x_fdr.RData") thresh<-GOthresh(x_m,x_fdr$sampleFDR,enrichThresh=2,countThresh=2,fdrThresh=0.100)
GoMiner data set
data(Housekeeping_Genes)data(Housekeeping_Genes)
retrieve n unique random genes
randSubsetGeneList(geneList, ngenes)randSubsetGeneList(geneList, ngenes)
geneList |
character vector geneList |
ngenes |
integer desired number of random genes |
returns a character vector of genes
#load("data/GOGOA3small.RData") genes<-randSubsetGeneList(GOGOA3small$genes[["biological_process"]],20)#load("data/GOGOA3small.RData") genes<-randSubsetGeneList(GOGOA3small$genes[["biological_process"]],20)
prepare a cpd of p values from randomized gene sets
RCPD(GOGOA3, ngenes, nrand, ONT)RCPD(GOGOA3, ngenes, nrand, ONT)
GOGOA3 |
return value of subsetGOGOA() |
ngenes |
integer number of genes to randomize |
nrand |
integer number of randomizations |
ONT |
c("molecular_function","cellular_component","biological_process") |
the cpd of the randomizations is to be used for estimating the false discovery rate (FDR) of the real sampled genes
returns a histogram of log10(p)
#load("data/GOGOA3small.RData") rcpd<-RCPD(GOGOA3small,ngenes=100,nrand=10,ONT="biological_process")#load("data/GOGOA3small.RData") rcpd<-RCPD(GOGOA3small,ngenes=100,nrand=10,ONT="biological_process")
convert outdated HGNC symbols to current HGNC symbols
validHGNCSymbols(geneList)validHGNCSymbols(geneList)
geneList |
character vector of HGNC symbols |
removes NA and /// from output of checkGeneSymbols()
returns list of mapping table and vector of current HGNC symbols
geneList<-c("FN1", "tp53", "UNKNOWNGENE","7-Sep", "9/7", "1-Mar", "Oct4", "4-Oct","OCT4-PG4", "C19ORF71", "C19orf71") l<-validHGNCSymbols(geneList)geneList<-c("FN1", "tp53", "UNKNOWNGENE","7-Sep", "9/7", "1-Mar", "Oct4", "4-Oct","OCT4-PG4", "C19ORF71", "C19orf71") l<-validHGNCSymbols(geneList)