getHOMOLOG: Find homologous/orthologous gene (ID)

In alexandremkuhn/annotationTools: Annotate microarrays and perform cross-species gene expression analyses using flat file databases

Description

Takes a vector of gene IDs, a table of homologs/orthologs, and a target species and returns gene IDs corresponding to homologous/orthologous genes.

Usage

getHOMOLOG(geneid, targetspecies, homol, cluster = FALSE, diagnose = FALSE, noIDsymbol = NA, clusterCol = 1, speciesCol = 2, idCol = 3, tableType = "homologene")

Arguments

geneid	character vector containing gene IDs.
targetspecies	identifier of the target species in the homology/orthology table.
homol	homology/orthology table (data frame) listing gene IDs (1 per line) along with the species and the homology/orthology cluster they belong to.
cluster	logical. If TRUE, the identifiers provided in 'geneid' are homology/orthology cluster IDs. If FALSE, they are gene IDs.
diagnose	logical. If TRUE, 3 (logical) vectors used for diagnostic purpose are returned in addition to the annotation. If FALSE (default) only the annotation is returned.
noIDsymbol	character string to be used in output list 'targetid' if no homologous/orthologous gene is found or provided in the annotation table.
clusterCol	column in homology/orthology table containing homology/orthology cluster IDs.
speciesCol	column in homology/orthology table containing species IDs.
idCol	column in homology/orthology table containing gene IDs.
tableType	character string specifying the type of homology/orthology table used. Either 'homologene' (default) or 'gene_orthologs'.

Details

The homology/orthology table lists gene IDs (from several species) and the homology/orthology cluster they belong to. Homologous and orthologous genes share a common cluster identifier. Given a certain gene ID, a target species, and a homology/orthology table, all gene IDs belonging to the same homology/orthology cluster and to the specified target species are returned. Various homology/orthology databases can be used, in particular NCBI's HomoloGene and their 'Orthologs from Annotation pipeline', referred to as 'gene_orthologs' database (see details below). If 'targetspecies' is the species 'geneid' belongs to, by definition, homologous genes are returned (if listed). On the contrary, specifying a 'targetspecies' different from the host species 'geneid' belongs to, results in orthologous genes to be returned. Note that each gene ID is assumed to be unique and to belong to a single homology/orthology cluster.

Gene IDs of homologous/orthologous genes are returned as elements of list 'targetid'. If multiple (homologous/orthologous) gene IDs are provided for 'geneid[i]', a vector containing all gene IDs is returned as the 'i-th' element of list 'targetid'.

Default values for 'clusterCol', 'speciesCol', and 'idCol' are chosen to match the table provided by HomoloGene (homologene.data provided by www.ncbi.nlm.nih.gov/HomoloGene). Homology/orthology tables from other sources might require setting these values appropriately.

Orthologs defined in NCBI's 'Orthologs from Annotation pipeline' database (available at ftp.ncbi.nlm.nih.gov/gene/DATA/gene_orthologs.gz, and hence referred to as 'gene_orthologs') can be mined by setting 'tableType' to 'gene_orthologs' instead of 'homologene' (default). In this case, arguments 'clusterCol', 'speciesCol',and 'idCol' are overridden to fit the data structure used in 'gene_orthologs'. In short, one difference between 'gene_orthologs' compared to HomoloGene is that 'gene_orthologs' does not use ortholog cluster IDs but anchors each ortholog groups using the human gene ID of the ortholog group. If a specific ortholog group does not contain a human gene, a gene ID from another species within the ortholog group may be used as anchor.

'gene_orthologs' is a rich source of homologs/orthologs between selected vertebrate species and introduced by NCBI in 2014 (https://www.ncbi.nlm.nih.gov/kis/info/how-are-orthologs-calculated/). Note that 'gene_orthologs' does not list (and thus cannot be used to search for) homologs (i.e. only 'Ortholog' relationships, as specified in the 3rd column of 'gene_orthologs' are considered).

Finally, if 'cluster' is TRUE, cluster IDs can be provided in 'geneid' (instead of gene IDs) and the function will return all (homologous/orthologous) gene IDs belonging to a given cluster ID and a given 'targetspecies'. This can be used to mine orthology tables provided by Affymetrix (e.g. 'Mouse430\_2\_ortholog.csv') for orthologous probe set IDs (see 'examples' below).

Value

targetid	list of length 'length(geneid)' the 'i'-th element of which contains the homologous/orthologous gene IDs for 'geneid[i]' and 'targetspecies'.
empty	logical vector of length 'length(geneid)'. 'empty[i]' is TRUE if 'geneid[i]' is empty or NA.
noentry	locial vector of length 'length(geneid)'. 'noentry[i]' is TRUE if 'geneid[i]' cannot be found in column 'idCol' (default is column 3) of the homology/orthology table 'homol'.
notargetid	locial vector of length 'length(geneid)'. 'notargetid[i]' is TRUE if 'geneed[i]' is found in the homology/orthology table but no homolog/ortholog is listed for 'targetspecies'.

Author(s)

Alexandre Kuhn

References

Kuhn et al. Cross-species and cross-platform gene expression studies with the Bioconductor-compliant R package 'annotationTools'. BMC Bioinformatics, 9:26 (2008)

Examples

##example Homologene file and its location
homologeneFile<-system.file('extdata','homologene_part.data',package='annotationTools')

##load Homologene file
homologene<-read.delim(homologeneFile,header=FALSE)

##get mouse (species ID 10090) orthologs of several human (species ID 9606) gene ID (among those: 5982, gene symbol RFC2 and 93587, gene symbol: RG9MTD2)
myGenes<-c(5982,93587,NA,100)
getHOMOLOG(myGenes,10090,homologene)

##track origin of annotation failure for the last 2 gene IDs
getHOMOLOG(myGenes,10090,homologene,diagnose=TRUE)

##get mouse gene belonging to homologene cluster IDs 6885 and 6886
myClusters<-c(6885,6886)
getHOMOLOG(myClusters,10090,homologene,cluster=TRUE)

##get mouse orthologs of human genes using 'gene_orthologs'
gene_orthologsFile<-system.file('extdata','gene_orthologs_part.data',package='annotationTools')
gene_orthologs<-read.delim(gene_orthologsFile,header=TRUE)
getHOMOLOG(myGenes,10090,gene_orthologs,tableType='gene_orthologs')

##mine Affymetrix (example) ortholog file
affyOrthologFile<-system.file('extdata','HG-U133_Plus_2_ortholog_part.csv',package='annotationTools')
affyOrthologs<-read.csv(affyOrthologFile,colClasses='character')

##get Mouse430_2 probe set IDs 'orthologous' to HG-U133_Plus_2 probe set IDs 1053_at and 121_at
myPS<-c('1053_at','121_at')
getHOMOLOG(myPS,'Mouse430_2',affyOrthologs,cluster=TRUE,clusterCol=1,speciesCol=4,idCol=3)

alexandremkuhn/annotationTools documentation built on April 5, 2020, 12:59 a.m.