I am using minfi to analyze some illumina 450K arrays. I have a list of differentially methylated positions that I want to convert to gene names, or something useful. There's no information on annotation in the minfi manual, but I can get the sequence from 'IlluminaHumanMethylation450kmanifest'. I found the pdf "Resolve the inconsistency of Illumina identifiers through nuID", which sounds like it should help me do what want. If I understand correctly, all I have to do is convert my sequence to a nuID, and then conver the nuID to a refseq or gene symbol or what have you. I can do the first part easily: >ProbeInfo <- getProbeInfo(IlluminaHumanMethylation450kmanifest, type ="I") > head(dmp, n=1) intercept f pval qval cg07466271 2.232301 39.90621 1.447446e-09 0.0006144155 > which(ProbeInfo$Name == "cg07466271") [1] 77780 > ProbeInfo[77780,] Name AddressA AddressB Color NextBase 293109 cg07466271 49719346 67746443 Red A ProbeSeqA 293109 TAACCCAAACCCCACTTCAAATACACAACAAATTCCTTATAAAAACCTCA ProbeSeqB nCpG 293109 TAACCCGAACCCCGCTTCAAATACACAACAAATTCCTTATAAAAACCTCG 2 > seq2id("TAACCCAAACCCCACTTCAAATACACAACAAATTCCTTATAAAAACCTCA") [1] "ZwVAVR9AxEED18wAXQ" Great, now I have my nuID. The problem occurs when I try to convert this into something useful: > library(lumiHumanIDMapping) > if(require(lumiHumanIDMapping)) nuID2RefSeqID("ZwVAVR9AxEED18wAXQ", lib.mapping='lumiHumanIDMapping') The provided names of filterTh does not match the field names of nuID_MappingInfo table. No filtering will be performed. ZwVAVR9AxEED18wAXQ NA Warning message: In getNuIDMappingInfo(nuID, lib.mapping = lib.mapping) : No matches were found! I also tried it with 'lumiHumanAll.db', and got the same result: > library(annotate) > getSYMBOL("ZwVAVR9AxEED18wAXQ", 'lumiHumanAll.db') ZwVAVR9AxEED18wAXQ NA I get the same thing if I try it with the "IlluminaHumanMethylation450k.db" package: > library("IlluminaHumanMethylation450k.db") > getSYMBOL("ZwVAVR9AxEED18wAXQ", 'IlluminaHumanMethylation450k.db') ZwVAVR9AxEED18wAXQ NA What's going on here?

I would suggest that you use the approach I outlined in my earlier email, but here is how the 450k.db sequences work: library(lumi) library(IlluminaHumanMethylation450k.db) id2seq(IlluminaHumanMethylation450kNUID[['cg07466271']]) ## [1] "TGGCCCGGACCCCGCTTCAGGTGCACAGCAAATTCCTTGTAGGAGCCTCG" In terms of mapping probes to genes, I STRONGLY suggest that you use a range-based approach and tailor it to your intended goal. Pan and I discussed NuID identifiers for the methylation arrays, and he felt it was too much trouble. Kasper notes in the minfi vignette that the cgXXX identifier in Illumina's probe names is followed by a hash of the (source? forward?) sequence and is therefore unique. That obviates the primary purpose of using NuID for expression arrays, namely, the genes "moving around" from assembly to assembly. --t On Fri, May 4, 2012 at 2:26 PM, Ed Siefker <ebs15242@gmail.com> wrote: > I am using minfi to analyze some illumina 450K arrays. > I have a list of differentially methylated positions that > I want to convert to gene names, or something useful. > > There's no information on annotation in the minfi > manual, but I can get the sequence from > 'IlluminaHumanMethylation450kmanifest'. > > I found the pdf "Resolve the inconsistency of > Illumina identifiers through nuID", which sounds > like it should help me do what want. > > If I understand correctly, all I have to do is convert my > sequence to a nuID, and then conver the nuID to a > refseq or gene symbol or what have you. > > I can do the first part easily: > > >ProbeInfo <- getProbeInfo(IlluminaHumanMethylation450kmanifest, type ="I") > > head(dmp, n=1) > intercept f pval qval > cg07466271 2.232301 39.90621 1.447446e-09 0.0006144155 > > > which(ProbeInfo$Name == "cg07466271") > [1] 77780 > > > ProbeInfo[77780,] > Name AddressA AddressB Color NextBase > 293109 cg07466271 49719346 67746443 Red A > ProbeSeqA > 293109 TAACCCAAACCCCACTTCAAATACACAACAAATTCCTTATAAAAACCTCA > ProbeSeqB nCpG > 293109 TAACCCGAACCCCGCTTCAAATACACAACAAATTCCTTATAAAAACCTCG 2 > > > seq2id("TAACCCAAACCCCACTTCAAATACACAACAAATTCCTTATAAAAACCTCA") > [1] "ZwVAVR9AxEED18wAXQ" > > > Great, now I have my nuID. The problem occurs when I try to convert this > into something useful: > > > library(lumiHumanIDMapping) > > if(require(lumiHumanIDMapping)) nuID2RefSeqID("ZwVAVR9AxEED18wAXQ", > lib.mapping='lumiHumanIDMapping') > The provided names of filterTh does not match the field names of > nuID_MappingInfo table. > No filtering will be performed. > ZwVAVR9AxEED18wAXQ > NA > Warning message: > In getNuIDMappingInfo(nuID, lib.mapping = lib.mapping) : > No matches were found! > > > I also tried it with 'lumiHumanAll.db', and got the same result: > > > library(annotate) > > getSYMBOL("ZwVAVR9AxEED18wAXQ", 'lumiHumanAll.db') > ZwVAVR9AxEED18wAXQ > NA > > > I get the same thing if I try it with the "IlluminaHumanMethylation450k.db" > package: > > > > library("IlluminaHumanMethylation450k.db") > > getSYMBOL("ZwVAVR9AxEED18wAXQ", 'IlluminaHumanMethylation450k.db') > ZwVAVR9AxEED18wAXQ > NA > > > > What's going on here? > > _______________________________________________ > Bioconductor mailing list > Bioconductor@r-project.org > https://stat.ethz.ch/mailman/listinfo/bioconductor > Search the archives: > http://news.gmane.org/gmane.science.biology.informatics.conductor > -- *A model is a lie that helps you see the truth.* * * Howard Skipper<http: cancerres.aacrjournals.org="" content="" 31="" 9="" 1173.full.pdf=""> [[alternative HTML version deleted]]