Question about using oligo and SNPchip for copy number analysis
Entering edit mode
Min-Han Tan ▴ 40
Last seen 7.7 years ago
Good morning, I apologize for troubling the list. I have researched this issue for a day, and spent some time troubleshooting but am unable to locate the answer. I am using R 2.7.2, Bioconductor 2.2 with oligo package version 1.4 and SNPchip 1.4. I am new to high throughput SNP analysis. My goal is to perform copy number / LOH analysis on tumor tissue samples processed with Affymetrix 100K SNP chips. Question : 1. My understanding is that I need an oligoSNPset class object to work on in SNPchip. How do I use oligo to generate an oligoSNPset class object? (genotypes + copy number). In a related 2nd question, I need to merge the oligoSNPset class objects from Xba and Hind CEL files for 100K Affymetrix chips - I can only find online how to do that for SnpCallSetPLUS class objects. 2. How may I merge the 2 separate SnpQset class objects from justSNPRMA (Hind and Xba chips) from dual chip Affymetrix 100K SNP chips? How may I use that object in SNPchip for copy number analysis, if possible? DETAILS: 1. I have no problems with the vignette in SNPchip, and viewing the plot. > library(SNPchip) > data(sample.snpset) > sample.snpset > snpset <- sample.snpset[chromosome(sample.snpset) =="1",1] > graph.par <- plotSnp(snpset[,1]) > graph.par$label.cytoband <- FALSE > graph.par$use.chromosome.size <- TRUE > graph.par Now sample.snpset belongs to class oligoSnpSet. I tried, but can only find out how to create objects of SnpCallSetPlus class (genotype calls) from CEL files, using combine to merge Xba and Hind results. > library(oligo); library(hapmap100kxba); library(hapmap100khind) ... > xba <- justCRLMM(fullFilenames, phenoData = pd,verbose = FALSE) > hind <- justCRLMM(fullFilenames, phenoData = pd,verbose = FALSE) ... > callset <- combine(xba,hind) > str(callset) Formal class 'SnpCallSetPlus' [package "oligoClasses"] with 6 slots I have looked at the PDF file in oligoclasses showing that SnpCallSet belongs to a different class (genotype calls) compared to oligoSnpSet (genotype calls and copy number estimates), but I am still uncertain, I am quite new to this sort of analysis. I have read a comment in the oligo vignette oHowTo.pdfdirecting me to the inst/testing directory of VanillaICE where there is apparently some information on the analysis of 209 CEL files using VanillaICE, but I am unable to locate the relevant file. 2. If CRLMM is not going to be possible, secondly, how may I merge the output of SNPRMA, and use it for the same analysis? I read that the SNPRMA algorithm summarizes the results, but I am unable to merge the Xba and Hind files as they stand. > xba.rma <- justSNPRMA(fullFilenames, verbose = TRUE) > hind.rma <- justSNPRMA(fullFilenames, verbose = TRUE) > combine(xba.rma,hind.rma) Error in combine(xba.rma, hind.rma) : objects have different annotations: pd.mapping50k.xba240, pd.mapping50k.hind240 > str(xba.rma) Formal class 'SnpQSet' [package "oligoClasses"] with 6 slots Your advice is much appreciated! I am sorry if this seems obvious. Min-han / Min-Han Tan, MBBS, MRCP Associate Consultant Department of Medical Oncology National Cancer Centre 11 Hospital Drive, Singapore 169610 Republic of Singapore Tel : +65 64368000 Email : [[alternative HTML version deleted]]
SNP Cancer graph oligo SNPchip oligoClasses VanillaICE crlmm SNP Cancer graph oligo • 916 views

Login before adding your answer.

Traffic: 175 users visited in the last hour
Help About
Access RSS

Use of this site constitutes acceptance of our User Agreement and Privacy Policy.

Powered by the version 2.3.6