Thanks, that did the trick. Workflow finished as expected. One question, though, to fully understand: what is the exact meaning of .*$ in the argument pattern? I tried to look it up but only found that: " * The preceding item will be matched zero or more times. " Thanks, Guido -----Original Message----- From: Ryan C. Thompson [mailto:rct@thompsonclan.org] Sent: Thursday, February 28, 2013 23:24 To: Hooiveld, Guido Cc: bioconductor at r-project.org Subject: Re: [BioC] annotationTools: character vector clean-up You can try this: library(stringr) x <- str_replace(string=x, pattern=" /// .*$", replacement="") stopifnot(!any(str_detect(x, "///")) You might want to adjust the pattern to allow arbitrary spacing rather than just single spaces. On Thu 28 Feb 2013 02:11:21 PM PST, Hooiveld, Guido wrote: > Hi, > I have a simple problem that's driving me nuts... Any hints are appreciated! > > I am retrieving the human homologues of rat genes. I use the functions 'getHOMOLOG' and 'listToCharacterVector' from the library annotationTools. Everything is going fine, except for one thing: > Some rows (genes) contain multiple entries (homologues); for such row I would like to get rid of all entries except the first one. > Example: for row 18634 I currently have: > [18634] "6173 /// 100529097" > > I would like to get rid of everything except the first entry, so to get this: > [18634] "6173" > > How to do this for all relevant rows? Basically, I thus would like to remove everything positioned after the first number, starting with space-3xforwardslash-etc. > Thanks, > Guido > > > library(annotationTools) > library(hugene11stv1hsentrezg.db) > library(ragene11stv1rnentrezg.db) > > #Download HomoloGene data from: > #ftp://ftp.ncbi.nih.gov/pub/HomoloGene/current/ > homologene<-read.delim("homologene.data.121212.data",header=FALSE) # > (date of file manually added to name when saving download) colnames > (homologene) <- c ("HomologyGroupID", "TaxonID", "EgID", "Symbol", > "ProteinGI", "ProteinAcc") > > # Read rat probesets that are on the array as Entrez IDs; this returns > a list which is converted to a character vector # Next the probesets > that don't have an EntrezID are removed rat.eg.array <- > mget(ls(ragene11stv1rnentrezgENTREZID), ragene11stv1rnentrezgENTREZID) > rat.eg.array <- listToCharacterVector(rat.eg.array) > rat.eg.array <- rat.eg.array[!is.na(rat.eg.array)] > > # Convert rat EG IDs into human (9606) homologs; this returns a list > which is converted to a character vector >> rat2human <- getHOMOLOG(rat.eg.array,9606,homologene) #this takes >> some time > Warning messages: > 1: In getHOMOLOG(rat.eg.array, 9606, homologene) : > One or more gene input gene ID/cluster not found in homologue table > 2: In getHOMOLOG(rat.eg.array, 9606, homologene) : > One or more gene ID/cluster with no target provided in homologue > table >> rat2human <- listToCharacterVector(rat2human) >> class(rat2human) > [1] "character" >> >> head(rat2human) > [1] "54552" "80212" "11277" "10663" "199692" "399947" >> >> #example of multiple entries >> rat2human[18634] > [1] "6173 /// 100529097" >> > > > > --------------------------------------------------------- > Guido Hooiveld, PhD > Nutrition, Metabolism & Genomics Group Division of Human Nutrition > Wageningen University Biotechnion, Bomenweg 2 > NL-6703 HD Wageningen > the Netherlands > tel: (+)31 317 485788 > fax: (+)31 317 483342 > email: guido.hooiveld at wur.nl > internet: http://nutrigene.4t.com > http://scholar.google.com/citations?user=qFHaMnoAAAAJ > http://www.researcherid.com/rid/F-4912-2010 > > > [[alternative HTML version deleted]] > > _______________________________________________ > Bioconductor mailing list > Bioconductor at r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor