methyAnalysis for DMR identification
1
0
Entering edit mode
Guest User ★ 13k
@guest-user-4897
Last seen 10.2 years ago
Hello, I was wondering if anyone here has tried methyAnalysis for identifying Differentially Methylated Regions in Illumina 450k array data. It seems to have wonderful data-visualization options, but it found no differentially-methylated regions in my data (56 450k data sets), which was surprising. I've found very little documentation for this method, and I am wondering if I did something wrong - for example, should I have performed the smoothMethyData function separately, with a different window size? Relevant code snippets and output are provided below. I'd greatly appreciate any thoughts or suggestions. Thank you very much! Allegra __________________________________________________________ Code: library(lumi); library(methylumi); library(wateRmelon); library(methyAnalysis); library(IlluminaHumanMethylation450k.db); library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be installed manually barcodes <- readLines("barcodes.txt"); # read barcodes from a text file classes <- readLines("IdatClasses.txt"); data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) *** additional data processing *** data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object data.g <- MethyLumiM2GenoSet(data.m, lib = "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data and find DMRs allDMRInfo <- identifySigDMR(dmrResult); ___________________________________________________________________ Standard Output: 65 probes were removed because of lack of chromosome location information! Smoothing Chromosome chr1 ... Smoothing Chromosome chr10 ... Smoothing Chromosome chr11 ... Smoothing Chromosome chr12 ... Smoothing Chromosome chr13 ... Smoothing Chromosome chr14 ... Smoothing Chromosome chr15 ... Smoothing Chromosome chr16 ... Smoothing Chromosome chr17 ... Smoothing Chromosome chr18 ... Smoothing Chromosome chr19 ... Smoothing Chromosome chr2 ... Smoothing Chromosome chr20 ... Smoothing Chromosome chr21 ... Smoothing Chromosome chr22 ... Smoothing Chromosome chr3 ... Smoothing Chromosome chr4 ... Smoothing Chromosome chr5 ... Smoothing Chromosome chr6 ... Smoothing Chromosome chr7 ... Smoothing Chromosome chr8 ... Smoothing Chromosome chr9 ... Smoothing Chromosome chrX ... Smoothing Chromosome chrY ... [1] "Done with detectDMR.slideWin\n" No significant CpG-sites were identified based on current criteria! -- output of sessionInfo(): R version 3.0.2 (2013-09-25) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C [3] LC_TIME=en_US.utf8 LC_COLLATE=C [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 [7] LC_PAPER=en_US.utf8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C attached base packages: [1] grid parallel methods stats graphics grDevices utils [8] datasets base other attached packages: [1] MASS_7.3-29 [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 [3] minfi_1.9.11 [4] bumphunter_1.2.0 [5] locfit_1.5-9.1 [6] iterators_1.0.6 [7] foreach_1.4.1 [8] Biostrings_2.30.1 [9] lattice_0.20-24 [10] methyAnalysis_1.4.2 [11] GenomicRanges_1.14.4 [12] XVector_0.2.0 [13] IRanges_1.20.6 [14] wateRmelon_1.2.2 [15] ROC_1.38.0 [16] IlluminaHumanMethylation450k.db_2.0.7 [17] org.Hs.eg.db_2.10.1 [18] RSQLite_0.11.4 [19] DBI_0.2-7 [20] AnnotationDbi_1.24.0 [21] limma_3.18.4 [22] methylumi_2.8.0 [23] matrixStats_0.8.14 [24] ggplot2_0.9.3.1 [25] reshape2_1.2.2 [26] scales_0.2.3 [27] lumi_2.14.1 [28] Biobase_2.22.0 [29] BiocGenerics_0.8.0 loaded via a namespace (and not attached): [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 [55] xtable_1.7-1 zlibbioc_1.8.0 ------------------------------------------------------------ Sender: LSF System <lsfadmin at="" blade13-2-16.gsc.wustl.edu=""> Subject: Job 5342893: <idat2dmrma> Done -- Sent via the guest posting facility at bioconductor.org.
convert methyAnalysis convert methyAnalysis • 2.5k views
ADD COMMENT
0
Entering edit mode
@giovanni-calice-6415
Last seen 8 months ago
Italy
Hi Allegra, I'm having some difficulties in identifying Differentially Methylated Regions on my datasets too, both with bumphunter/minfi and with methyAnalysis package. With regard to bumphunter/minfi there was a clarification also to me in the previous massages of the BioC list. With regard to methyAnalysis I think you performed the smoothMethyData function separately It isn't wrong; in fact, if you had not smoothing separately, the next method of the pipeline detectDMR.slideWin performs it first and then proceeds with differential analysis. According to me you could try setting a different p.value.detection.thargument and try to see if you get results. Instead I get results in identifying Differentially Methylated Regions by methyAnalysis but none is significant (p.adjust = 1 or p.adjust > 0.7). So I'm wondering could the small number of samples (I've 8-12 samples for now) affect the significance of results and lead to a p.adjust not so good? Best, Giovanni Laboratory of Preclinical and Translational Research IRCCS - CROB Oncology Referral Center of Basilicata Rionero in Vulture - Italy *** Mail from ************************ Bioconductor Digest, Vol 132, Issue 26 ************************************** ************************************** Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) From: "Allegra Petti [guest]" <guest@bioconductor.org> To: bioconductor@r-project.org, apetti@genome.wustl.edu Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> Subject: [BioC] methyAnalysis for DMR identification Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> Hello, I was wondering if anyone here has tried methyAnalysis for identifying Differentially Methylated Regions in Illumina 450k array data. It seems to have wonderful data-visualization options, but it found no differentially-methylated regions in my data (56 450k data sets), which was surprising. I've found very little documentation for this method, and I am wondering if I did something wrong - for example, should I have performed the smoothMethyData function separately, with a different window size? Relevant code snippets and output are provided below. I'd greatly appreciate any thoughts or suggestions. Thank you very much! Allegra __________________________________________________________ Code: library(lumi); library(methylumi); library(wateRmelon); library(methyAnalysis); library(IlluminaHumanMethylation450k.db); library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be installed manually barcodes <- readLines("barcodes.txt"); # read barcodes from a text file classes <- readLines("IdatClasses.txt"); data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) *** additional data processing *** data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object data.g <- MethyLumiM2GenoSet(data.m, lib = "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data and find DMRs allDMRInfo <- identifySigDMR(dmrResult); ___________________________________________________________________ Standard Output: 65 probes were removed because of lack of chromosome location information! Smoothing Chromosome chr1 ... Smoothing Chromosome chr10 ... Smoothing Chromosome chr11 ... Smoothing Chromosome chr12 ... Smoothing Chromosome chr13 ... Smoothing Chromosome chr14 ... Smoothing Chromosome chr15 ... Smoothing Chromosome chr16 ... Smoothing Chromosome chr17 ... Smoothing Chromosome chr18 ... Smoothing Chromosome chr19 ... Smoothing Chromosome chr2 ... Smoothing Chromosome chr20 ... Smoothing Chromosome chr21 ... Smoothing Chromosome chr22 ... Smoothing Chromosome chr3 ... Smoothing Chromosome chr4 ... Smoothing Chromosome chr5 ... Smoothing Chromosome chr6 ... Smoothing Chromosome chr7 ... Smoothing Chromosome chr8 ... Smoothing Chromosome chr9 ... Smoothing Chromosome chrX ... Smoothing Chromosome chrY ... [1] "Done with detectDMR.slideWin\n" No significant CpG-sites were identified based on current criteria! -- output of sessionInfo(): R version 3.0.2 (2013-09-25) Platform: x86_64-unknown-linux-gnu (64-bit) locale: [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C [3] LC_TIME=en_US.utf8 LC_COLLATE=C [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 [7] LC_PAPER=en_US.utf8 LC_NAME=C [9] LC_ADDRESS=C LC_TELEPHONE=C [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C attached base packages: [1] grid parallel methods stats graphics grDevices utils [8] datasets base other attached packages: [1] MASS_7.3-29 [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 [3] minfi_1.9.11 [4] bumphunter_1.2.0 [5] locfit_1.5-9.1 [6] iterators_1.0.6 [7] foreach_1.4.1 [8] Biostrings_2.30.1 [9] lattice_0.20-24 [10] methyAnalysis_1.4.2 [11] GenomicRanges_1.14.4 [12] XVector_0.2.0 [13] IRanges_1.20.6 [14] wateRmelon_1.2.2 [15] ROC_1.38.0 [16] IlluminaHumanMethylation450k.db_2.0.7 [17] org.Hs.eg.db_2.10.1 [18] RSQLite_0.11.4 [19] DBI_0.2-7 [20] AnnotationDbi_1.24.0 [21] limma_3.18.4 [22] methylumi_2.8.0 [23] matrixStats_0.8.14 [24] ggplot2_0.9.3.1 [25] reshape2_1.2.2 [26] scales_0.2.3 [27] lumi_2.14.1 [28] Biobase_2.22.0 [29] BiocGenerics_0.8.0 loaded via a namespace (and not attached): [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 [55] xtable_1.7-1 zlibbioc_1.8.0 ************************************** ************************************** [[alternative HTML version deleted]]
ADD COMMENT
0
Entering edit mode
Hi Allegra and Giovanni The parameters of identifySigDMR need to be tuned based on your dataset. The default parameters was designed for the cell lines with clear difference between groups. For tissue samples, these parameters need to be less stringent, especially the diffTh parameter. Also, I strongly recommend visually checking your data. You can export the methylation data using export.methyGenoSet, and then visualize it in IGV. Pan On Tue, Feb 25, 2014 at 8:31 AM, Giovanni Calice <giovcalice@gmail.com>wrote: > Hi Allegra, > > I'm having some difficulties in identifying Differentially Methylated > Regions on my datasets too, > both with bumphunter/minfi and with methyAnalysis package. > > With regard to bumphunter/minfi there was a clarification also to me > in the previous massages of the BioC list. > > With regard to methyAnalysis I think you performed the smoothMethyData > function separately > It isn't wrong; in fact, if you had not smoothing separately, the next > method of the pipeline detectDMR.slideWin > performs it first and then proceeds with differential analysis. > > According to me you could try setting a different p.value.detection.thargument and try to see if you get results. > > Instead I get results in identifying Differentially Methylated Regions by > methyAnalysis but > none is significant (p.adjust = 1 or p.adjust > 0.7). > > So I'm wondering could the small number of samples (I've 8-12 samples for > now) affect > the significance of results and lead to a p.adjust not so good? > > Best, > Giovanni > > > Laboratory of Preclinical and Translational Research > IRCCS - CROB Oncology Referral Center of Basilicata > Rionero in Vulture - Italy > > *** Mail from ************************ > Bioconductor Digest, Vol 132, Issue 26 > ************************************** > ************************************** > Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) > From: "Allegra Petti [guest]" <guest@bioconductor.org> > To: bioconductor@r-project.org, apetti@genome.wustl.edu > Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> > Subject: [BioC] methyAnalysis for DMR identification > Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> > > > > Hello, > > I was wondering if anyone here has tried methyAnalysis for identifying > Differentially Methylated Regions in Illumina 450k array data. It seems to > have wonderful data-visualization options, but it found no > differentially-methylated regions in my data (56 450k data sets), which was > surprising. I've found very little documentation for this method, and I am > wondering if I did something wrong - for example, should I have performed > the smoothMethyData function separately, with a different window size? > Relevant code snippets and output are provided below. I'd greatly > appreciate any thoughts or suggestions. > > Thank you very much! > Allegra > __________________________________________________________ > Code: > > library(lumi); > library(methylumi); > library(wateRmelon); > library(methyAnalysis); > library(IlluminaHumanMethylation450k.db); > library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be > installed manually > barcodes <- readLines("barcodes.txt"); # read barcodes from a text file > classes <- readLines("IdatClasses.txt"); > data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) > *** additional data processing *** > data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object > data.g <- MethyLumiM2GenoSet(data.m, lib = > "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet > dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data > and find DMRs > allDMRInfo <- identifySigDMR(dmrResult); > ___________________________________________________________________ > Standard Output: > > 65 probes were removed because of lack of chromosome location information! > Smoothing Chromosome chr1 ... > Smoothing Chromosome chr10 ... > Smoothing Chromosome chr11 ... > Smoothing Chromosome chr12 ... > Smoothing Chromosome chr13 ... > Smoothing Chromosome chr14 ... > Smoothing Chromosome chr15 ... > Smoothing Chromosome chr16 ... > Smoothing Chromosome chr17 ... > Smoothing Chromosome chr18 ... > Smoothing Chromosome chr19 ... > Smoothing Chromosome chr2 ... > Smoothing Chromosome chr20 ... > Smoothing Chromosome chr21 ... > Smoothing Chromosome chr22 ... > Smoothing Chromosome chr3 ... > Smoothing Chromosome chr4 ... > Smoothing Chromosome chr5 ... > Smoothing Chromosome chr6 ... > Smoothing Chromosome chr7 ... > Smoothing Chromosome chr8 ... > Smoothing Chromosome chr9 ... > Smoothing Chromosome chrX ... > Smoothing Chromosome chrY ... > [1] "Done with detectDMR.slideWin\n" > No significant CpG-sites were identified based on current criteria! > > > > -- output of sessionInfo(): > > R version 3.0.2 (2013-09-25) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C > [3] LC_TIME=en_US.utf8 LC_COLLATE=C > [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 > [7] LC_PAPER=en_US.utf8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C > > attached base packages: > [1] grid parallel methods stats graphics grDevices utils > [8] datasets base > > other attached packages: > [1] MASS_7.3-29 > [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 > [3] minfi_1.9.11 > [4] bumphunter_1.2.0 > [5] locfit_1.5-9.1 > [6] iterators_1.0.6 > [7] foreach_1.4.1 > [8] Biostrings_2.30.1 > [9] lattice_0.20-24 > [10] methyAnalysis_1.4.2 > [11] GenomicRanges_1.14.4 > [12] XVector_0.2.0 > [13] IRanges_1.20.6 > [14] wateRmelon_1.2.2 > [15] ROC_1.38.0 > [16] IlluminaHumanMethylation450k.db_2.0.7 > [17] org.Hs.eg.db_2.10.1 > [18] RSQLite_0.11.4 > [19] DBI_0.2-7 > [20] AnnotationDbi_1.24.0 > [21] limma_3.18.4 > [22] methylumi_2.8.0 > [23] matrixStats_0.8.14 > [24] ggplot2_0.9.3.1 > [25] reshape2_1.2.2 > [26] scales_0.2.3 > [27] lumi_2.14.1 > [28] Biobase_2.22.0 > [29] BiocGenerics_0.8.0 > > loaded via a namespace (and not attached): > [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 > [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 > [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 > [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 > [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 > [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 > [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 > [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 > [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 > [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 > [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 > [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 > [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 > [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 > 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 > [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 > [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 > [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 > [55] xtable_1.7-1 zlibbioc_1.8.0 > > ************************************** > ************************************** > [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Pan and Giovanni, Thanks for your replies, and for the advice on data visualization. I've found very little documentation for these functions, so I don't have a good feeling for the parameters and how they should be set. How is diffTh defined, and what is its range? Do lower diffTh values correspond to lower stringency? Do you have any tips for translating the data visualization into meaningful parameters? Thank so much! Allegra _________________________________ Allegra A. Petti, Ph.D. The Genome Institute 4444 Forest Park Ave. St. Louis, MO 63108 apetti@genome.wustl.edu allegra.conbrio@post.harvard.edu On Feb 25, 2014, at 11:23 AM, Pan Du <dupan.mail@gmail.com> wrote: > Hi Allegra and Giovanni > > The parameters of identifySigDMR need to be tuned based on your dataset. The default parameters was designed for the cell lines with clear difference between groups. For tissue samples, these parameters need to be less stringent, especially the diffTh parameter. > Also, I strongly recommend visually checking your data. You can export the methylation data using export.methyGenoSet, and then visualize it in IGV. > > Pan > > > > On Tue, Feb 25, 2014 at 8:31 AM, Giovanni Calice <giovcalice@gmail.com> wrote: > Hi Allegra, > > I'm having some difficulties in identifying Differentially Methylated Regions on my datasets too, > both with bumphunter/minfi and with methyAnalysis package. > > With regard to bumphunter/minfi there was a clarification also to me > in the previous massages of the BioC list. > > With regard to methyAnalysis I think you performed the smoothMethyData function separately > It isn't wrong; in fact, if you had not smoothing separately, the next method of the pipeline detectDMR.slideWin > performs it first and then proceeds with differential analysis. > > According to me you could try setting a different p.value.detection.th argument and try to see if you get results. > > Instead I get results in identifying Differentially Methylated Regions by methyAnalysis but > none is significant (p.adjust = 1 or p.adjust > 0.7). > > So I'm wondering could the small number of samples (I've 8-12 samples for now) affect > the significance of results and lead to a p.adjust not so good? > > Best, > Giovanni > > > Laboratory of Preclinical and Translational Research > IRCCS - CROB Oncology Referral Center of Basilicata > Rionero in Vulture - Italy > > *** Mail from ************************ > Bioconductor Digest, Vol 132, Issue 26 > ************************************** > ************************************** > Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) > From: "Allegra Petti [guest]" <guest@bioconductor.org> > To: bioconductor@r-project.org, apetti@genome.wustl.edu > Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> > Subject: [BioC] methyAnalysis for DMR identification > Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> > > > > Hello, > > I was wondering if anyone here has tried methyAnalysis for identifying Differentially Methylated Regions in Illumina 450k array data. It seems to have wonderful data-visualization options, but it found no differentially-methylated regions in my data (56 450k data sets), which was surprising. I've found very little documentation for this method, and I am wondering if I did something wrong - for example, should I have performed the smoothMethyData function separately, with a different window size? Relevant code snippets and output are provided below. I'd greatly appreciate any thoughts or suggestions. > > Thank you very much! > Allegra > __________________________________________________________ > Code: > > library(lumi); > library(methylumi); > library(wateRmelon); > library(methyAnalysis); > library(IlluminaHumanMethylation450k.db); > library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be installed manually > barcodes <- readLines("barcodes.txt"); # read barcodes from a text file > classes <- readLines("IdatClasses.txt"); > data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) > *** additional data processing *** > data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object > data.g <- MethyLumiM2GenoSet(data.m, lib = "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet > dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data and find DMRs > allDMRInfo <- identifySigDMR(dmrResult); > ___________________________________________________________________ > Standard Output: > > 65 probes were removed because of lack of chromosome location information! > Smoothing Chromosome chr1 ... > Smoothing Chromosome chr10 ... > Smoothing Chromosome chr11 ... > Smoothing Chromosome chr12 ... > Smoothing Chromosome chr13 ... > Smoothing Chromosome chr14 ... > Smoothing Chromosome chr15 ... > Smoothing Chromosome chr16 ... > Smoothing Chromosome chr17 ... > Smoothing Chromosome chr18 ... > Smoothing Chromosome chr19 ... > Smoothing Chromosome chr2 ... > Smoothing Chromosome chr20 ... > Smoothing Chromosome chr21 ... > Smoothing Chromosome chr22 ... > Smoothing Chromosome chr3 ... > Smoothing Chromosome chr4 ... > Smoothing Chromosome chr5 ... > Smoothing Chromosome chr6 ... > Smoothing Chromosome chr7 ... > Smoothing Chromosome chr8 ... > Smoothing Chromosome chr9 ... > Smoothing Chromosome chrX ... > Smoothing Chromosome chrY ... > [1] "Done with detectDMR.slideWin\n" > No significant CpG-sites were identified based on current criteria! > > > > -- output of sessionInfo(): > > R version 3.0.2 (2013-09-25) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C > [3] LC_TIME=en_US.utf8 LC_COLLATE=C > [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 > [7] LC_PAPER=en_US.utf8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C > > attached base packages: > [1] grid parallel methods stats graphics grDevices utils > [8] datasets base > > other attached packages: > [1] MASS_7.3-29 > [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 > [3] minfi_1.9.11 > [4] bumphunter_1.2.0 > [5] locfit_1.5-9.1 > [6] iterators_1.0.6 > [7] foreach_1.4.1 > [8] Biostrings_2.30.1 > [9] lattice_0.20-24 > [10] methyAnalysis_1.4.2 > [11] GenomicRanges_1.14.4 > [12] XVector_0.2.0 > [13] IRanges_1.20.6 > [14] wateRmelon_1.2.2 > [15] ROC_1.38.0 > [16] IlluminaHumanMethylation450k.db_2.0.7 > [17] org.Hs.eg.db_2.10.1 > [18] RSQLite_0.11.4 > [19] DBI_0.2-7 > [20] AnnotationDbi_1.24.0 > [21] limma_3.18.4 > [22] methylumi_2.8.0 > [23] matrixStats_0.8.14 > [24] ggplot2_0.9.3.1 > [25] reshape2_1.2.2 > [26] scales_0.2.3 > [27] lumi_2.14.1 > [28] Biobase_2.22.0 > [29] BiocGenerics_0.8.0 > > loaded via a namespace (and not attached): > [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 > [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 > [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 > [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 > [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 > [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 > [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 > [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 > [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 > [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 > [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 > [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 > [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 > [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 > 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 > [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 > [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 > [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 > [55] xtable_1.7-1 zlibbioc_1.8.0 > > ************************************** > ************************************** > ____ This email message is a private communication. The information transmitted, including attachments, is intended only for the person or entity to which it is addressed and may contain confidential, privileged, and/or proprietary material. Any review, duplication, retransmission, distribution, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is unauthorized by the sender and is prohibited. If you have received this message in error, please contact the sender immediately by return email and delete the original message from all computer systems. Thank you. [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Pan and Giovanni, I realized I have one more question. I'm currently using methyAnalysis starting from idat files, but I'd also like to be able to start from a tab-delimited matrix of beta values. I couldn't figure out how to work with (and annotate) beta values in any of the R packages that produce MethyLumiM objects. Do you have any procedures for this? Thank you again, Allegra _________________________________ Allegra A. Petti, Ph.D. The Genome Institute 4444 Forest Park Ave. St. Louis, MO 63108 apetti@genome.wustl.edu allegra.conbrio@post.harvard.edu On Feb 25, 2014, at 11:23 AM, Pan Du <dupan.mail@gmail.com> wrote: > Hi Allegra and Giovanni > > The parameters of identifySigDMR need to be tuned based on your dataset. The default parameters was designed for the cell lines with clear difference between groups. For tissue samples, these parameters need to be less stringent, especially the diffTh parameter. > Also, I strongly recommend visually checking your data. You can export the methylation data using export.methyGenoSet, and then visualize it in IGV. > > Pan > > > > On Tue, Feb 25, 2014 at 8:31 AM, Giovanni Calice <giovcalice@gmail.com> wrote: > Hi Allegra, > > I'm having some difficulties in identifying Differentially Methylated Regions on my datasets too, > both with bumphunter/minfi and with methyAnalysis package. > > With regard to bumphunter/minfi there was a clarification also to me > in the previous massages of the BioC list. > > With regard to methyAnalysis I think you performed the smoothMethyData function separately > It isn't wrong; in fact, if you had not smoothing separately, the next method of the pipeline detectDMR.slideWin > performs it first and then proceeds with differential analysis. > > According to me you could try setting a different p.value.detection.th argument and try to see if you get results. > > Instead I get results in identifying Differentially Methylated Regions by methyAnalysis but > none is significant (p.adjust = 1 or p.adjust > 0.7). > > So I'm wondering could the small number of samples (I've 8-12 samples for now) affect > the significance of results and lead to a p.adjust not so good? > > Best, > Giovanni > > > Laboratory of Preclinical and Translational Research > IRCCS - CROB Oncology Referral Center of Basilicata > Rionero in Vulture - Italy > > *** Mail from ************************ > Bioconductor Digest, Vol 132, Issue 26 > ************************************** > ************************************** > Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) > From: "Allegra Petti [guest]" <guest@bioconductor.org> > To: bioconductor@r-project.org, apetti@genome.wustl.edu > Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> > Subject: [BioC] methyAnalysis for DMR identification > Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> > > > > Hello, > > I was wondering if anyone here has tried methyAnalysis for identifying Differentially Methylated Regions in Illumina 450k array data. It seems to have wonderful data-visualization options, but it found no differentially-methylated regions in my data (56 450k data sets), which was surprising. I've found very little documentation for this method, and I am wondering if I did something wrong - for example, should I have performed the smoothMethyData function separately, with a different window size? Relevant code snippets and output are provided below. I'd greatly appreciate any thoughts or suggestions. > > Thank you very much! > Allegra > __________________________________________________________ > Code: > > library(lumi); > library(methylumi); > library(wateRmelon); > library(methyAnalysis); > library(IlluminaHumanMethylation450k.db); > library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be installed manually > barcodes <- readLines("barcodes.txt"); # read barcodes from a text file > classes <- readLines("IdatClasses.txt"); > data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) > *** additional data processing *** > data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object > data.g <- MethyLumiM2GenoSet(data.m, lib = "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet > dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data and find DMRs > allDMRInfo <- identifySigDMR(dmrResult); > ___________________________________________________________________ > Standard Output: > > 65 probes were removed because of lack of chromosome location information! > Smoothing Chromosome chr1 ... > Smoothing Chromosome chr10 ... > Smoothing Chromosome chr11 ... > Smoothing Chromosome chr12 ... > Smoothing Chromosome chr13 ... > Smoothing Chromosome chr14 ... > Smoothing Chromosome chr15 ... > Smoothing Chromosome chr16 ... > Smoothing Chromosome chr17 ... > Smoothing Chromosome chr18 ... > Smoothing Chromosome chr19 ... > Smoothing Chromosome chr2 ... > Smoothing Chromosome chr20 ... > Smoothing Chromosome chr21 ... > Smoothing Chromosome chr22 ... > Smoothing Chromosome chr3 ... > Smoothing Chromosome chr4 ... > Smoothing Chromosome chr5 ... > Smoothing Chromosome chr6 ... > Smoothing Chromosome chr7 ... > Smoothing Chromosome chr8 ... > Smoothing Chromosome chr9 ... > Smoothing Chromosome chrX ... > Smoothing Chromosome chrY ... > [1] "Done with detectDMR.slideWin\n" > No significant CpG-sites were identified based on current criteria! > > > > -- output of sessionInfo(): > > R version 3.0.2 (2013-09-25) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C > [3] LC_TIME=en_US.utf8 LC_COLLATE=C > [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 > [7] LC_PAPER=en_US.utf8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C > > attached base packages: > [1] grid parallel methods stats graphics grDevices utils > [8] datasets base > > other attached packages: > [1] MASS_7.3-29 > [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 > [3] minfi_1.9.11 > [4] bumphunter_1.2.0 > [5] locfit_1.5-9.1 > [6] iterators_1.0.6 > [7] foreach_1.4.1 > [8] Biostrings_2.30.1 > [9] lattice_0.20-24 > [10] methyAnalysis_1.4.2 > [11] GenomicRanges_1.14.4 > [12] XVector_0.2.0 > [13] IRanges_1.20.6 > [14] wateRmelon_1.2.2 > [15] ROC_1.38.0 > [16] IlluminaHumanMethylation450k.db_2.0.7 > [17] org.Hs.eg.db_2.10.1 > [18] RSQLite_0.11.4 > [19] DBI_0.2-7 > [20] AnnotationDbi_1.24.0 > [21] limma_3.18.4 > [22] methylumi_2.8.0 > [23] matrixStats_0.8.14 > [24] ggplot2_0.9.3.1 > [25] reshape2_1.2.2 > [26] scales_0.2.3 > [27] lumi_2.14.1 > [28] Biobase_2.22.0 > [29] BiocGenerics_0.8.0 > > loaded via a namespace (and not attached): > [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 > [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 > [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 > [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 > [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 > [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 > [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 > [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 > [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 > [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 > [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 > [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 > [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 > [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 > 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 > [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 > [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 > [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 > [55] xtable_1.7-1 zlibbioc_1.8.0 > > ************************************** > ************************************** > ____ This email message is a private communication. The information transmitted, including attachments, is intended only for the person or entity to which it is addressed and may contain confidential, privileged, and/or proprietary material. Any review, duplication, retransmission, distribution, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is unauthorized by the sender and is prohibited. If you have received this message in error, please contact the sender immediately by return email and delete the original message from all computer systems. Thank you. [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Allegra In order to get the MethyLumiM object, you must have the methylated and unmethylated probe intensities in your tab-delimited files. The beta- values were basically calculated from the methylated and unmethylated probe intensities. I recommend using .IDAT files if you have them because it includes all the information. As for your question about diffTh parameter, you can think it as the log2-foldChange threshold, which is similar as what expression analysis always do. There are couple of reasons why visually check the methylation data (without smoothing) is important. Here is just list a couple of them: 1. The location of DMRs, because methylation regulation of transcript expression is very much chromosome location dependent. DMRs near TSS and CpG-islands are more likely to be involved in negative regulation of transcript expression. 2. Check whether there is consistent behavior of nearby probes (if there are) because methylation pattern tends to be locally correlated. 3. Globally check the methylation difference across samples. This may help the quality assessment of the samples, or identify whether there global-hyper or hypo methylation events. Pan On Tue, Feb 25, 2014 at 9:47 AM, Allegra A. Petti <apetti@genome.wustl.edu>wrote: > Hi Pan and Giovanni, > > I realized I have one more question. I'm currently using methyAnalysis > starting from idat files, but I'd also like to be able to start from a > tab-delimited matrix of beta values. I couldn't figure out how to work with > (and annotate) beta values in any of the R packages that produce MethyLumiM > objects. Do you have any procedures for this? > > Thank you again, > Allegra > _________________________________ > Allegra A. Petti, Ph.D. > > The Genome Institute > 4444 Forest Park Ave. > St. Louis, MO 63108 > apetti@genome.wustl.edu > allegra.conbrio@post.harvard.edu > > On Feb 25, 2014, at 11:23 AM, Pan Du <dupan.mail@gmail.com> wrote: > > Hi Allegra and Giovanni > > The parameters of identifySigDMR need to be tuned based on your dataset. > The default parameters was designed for the cell lines with clear > difference between groups. For tissue samples, these parameters need to be > less stringent, especially the diffTh parameter. > Also, I strongly recommend visually checking your data. You can export the > methylation data using export.methyGenoSet, and then visualize it in IGV. > > Pan > > > > On Tue, Feb 25, 2014 at 8:31 AM, Giovanni Calice <giovcalice@gmail.com>wrote: > >> Hi Allegra, >> >> I'm having some difficulties in identifying Differentially Methylated >> Regions on my datasets too, >> both with bumphunter/minfi and with methyAnalysis package. >> >> With regard to bumphunter/minfi there was a clarification also to me >> in the previous massages of the BioC list. >> >> With regard to methyAnalysis I think you performed the smoothMethyData >> function separately >> It isn't wrong; in fact, if you had not smoothing separately, the next >> method of the pipeline detectDMR.slideWin >> performs it first and then proceeds with differential analysis. >> >> According to me you could try setting a different p.value.detection.thargument and try to see if you get results. >> >> Instead I get results in identifying Differentially Methylated Regions by >> methyAnalysis but >> none is significant (p.adjust = 1 or p.adjust > 0.7). >> >> So I'm wondering could the small number of samples (I've 8-12 samples for >> now) affect >> the significance of results and lead to a p.adjust not so good? >> >> Best, >> Giovanni >> >> >> Laboratory of Preclinical and Translational Research >> IRCCS - CROB Oncology Referral Center of Basilicata >> Rionero in Vulture - Italy >> >> *** Mail from ************************ >> Bioconductor Digest, Vol 132, Issue 26 >> ************************************** >> ************************************** >> Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) >> From: "Allegra Petti [guest]" <guest@bioconductor.org> >> To: bioconductor@r-project.org, apetti@genome.wustl.edu >> Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> >> Subject: [BioC] methyAnalysis for DMR identification >> Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> >> >> >> >> Hello, >> >> I was wondering if anyone here has tried methyAnalysis for identifying >> Differentially Methylated Regions in Illumina 450k array data. It seems to >> have wonderful data-visualization options, but it found no >> differentially-methylated regions in my data (56 450k data sets), which was >> surprising. I've found very little documentation for this method, and I am >> wondering if I did something wrong - for example, should I have performed >> the smoothMethyData function separately, with a different window size? >> Relevant code snippets and output are provided below. I'd greatly >> appreciate any thoughts or suggestions. >> >> Thank you very much! >> Allegra >> __________________________________________________________ >> Code: >> >> library(lumi); >> library(methylumi); >> library(wateRmelon); >> library(methyAnalysis); >> library(IlluminaHumanMethylation450k.db); >> library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be >> installed manually >> barcodes <- readLines("barcodes.txt"); # read barcodes from a text file >> classes <- readLines("IdatClasses.txt"); >> data.s <- methylumIDAT(barcodes); # read in idat files (methylumi >> function) >> *** additional data processing *** >> data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object >> data.g <- MethyLumiM2GenoSet(data.m, lib = >> "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet >> dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth >> data and find DMRs >> allDMRInfo <- identifySigDMR(dmrResult); >> ___________________________________________________________________ >> Standard Output: >> >> 65 probes were removed because of lack of chromosome location information! >> Smoothing Chromosome chr1 ... >> Smoothing Chromosome chr10 ... >> Smoothing Chromosome chr11 ... >> Smoothing Chromosome chr12 ... >> Smoothing Chromosome chr13 ... >> Smoothing Chromosome chr14 ... >> Smoothing Chromosome chr15 ... >> Smoothing Chromosome chr16 ... >> Smoothing Chromosome chr17 ... >> Smoothing Chromosome chr18 ... >> Smoothing Chromosome chr19 ... >> Smoothing Chromosome chr2 ... >> Smoothing Chromosome chr20 ... >> Smoothing Chromosome chr21 ... >> Smoothing Chromosome chr22 ... >> Smoothing Chromosome chr3 ... >> Smoothing Chromosome chr4 ... >> Smoothing Chromosome chr5 ... >> Smoothing Chromosome chr6 ... >> Smoothing Chromosome chr7 ... >> Smoothing Chromosome chr8 ... >> Smoothing Chromosome chr9 ... >> Smoothing Chromosome chrX ... >> Smoothing Chromosome chrY ... >> [1] "Done with detectDMR.slideWin\n" >> No significant CpG-sites were identified based on current criteria! >> >> >> >> -- output of sessionInfo(): >> >> R version 3.0.2 (2013-09-25) >> Platform: x86_64-unknown-linux-gnu (64-bit) >> >> locale: >> [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C >> [3] LC_TIME=en_US.utf8 LC_COLLATE=C >> [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 >> [7] LC_PAPER=en_US.utf8 LC_NAME=C >> [9] LC_ADDRESS=C LC_TELEPHONE=C >> [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C >> >> attached base packages: >> [1] grid parallel methods stats graphics grDevices utils >> [8] datasets base >> >> other attached packages: >> [1] MASS_7.3-29 >> [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 >> [3] minfi_1.9.11 >> [4] bumphunter_1.2.0 >> [5] locfit_1.5-9.1 >> [6] iterators_1.0.6 >> [7] foreach_1.4.1 >> [8] Biostrings_2.30.1 >> [9] lattice_0.20-24 >> [10] methyAnalysis_1.4.2 >> [11] GenomicRanges_1.14.4 >> [12] XVector_0.2.0 >> [13] IRanges_1.20.6 >> [14] wateRmelon_1.2.2 >> [15] ROC_1.38.0 >> [16] IlluminaHumanMethylation450k.db_2.0.7 >> [17] org.Hs.eg.db_2.10.1 >> [18] RSQLite_0.11.4 >> [19] DBI_0.2-7 >> [20] AnnotationDbi_1.24.0 >> [21] limma_3.18.4 >> [22] methylumi_2.8.0 >> [23] matrixStats_0.8.14 >> [24] ggplot2_0.9.3.1 >> [25] reshape2_1.2.2 >> [26] scales_0.2.3 >> [27] lumi_2.14.1 >> [28] Biobase_2.22.0 >> [29] BiocGenerics_0.8.0 >> >> loaded via a namespace (and not attached): >> [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 >> [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 >> [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 >> [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 >> [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 >> [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 >> [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 >> [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 >> [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 >> [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 >> [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 >> [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 >> [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 >> [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 >> 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 >> [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 >> [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 >> [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 >> [55] xtable_1.7-1 zlibbioc_1.8.0 >> >> ************************************** >> ************************************** >> > > > > ____ This email message is a private communication. The information > transmitted, including attachments, is intended only for the person or > entity to which it is addressed and may contain confidential, privileged, > and/or proprietary material. Any review, duplication, retransmission, > distribution, or other use of, or taking of any action in reliance upon, > this information by persons or entities other than the intended recipient > is unauthorized by the sender and is prohibited. If you have received this > message in error, please contact the sender immediately by return email and > delete the original message from all computer systems. Thank you. > [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Pan, Thank you very much for your helpful response. I really appreciate it! --Allegra _________________________________ Allegra A. Petti, Ph.D. The Genome Institute 4444 Forest Park Ave. St. Louis, MO 63108 apetti@genome.wustl.edu allegra.conbrio@post.harvard.edu On Feb 25, 2014, at 5:51 PM, Pan Du <dupan.mail@gmail.com> wrote: > Hi Allegra > > In order to get the MethyLumiM object, you must have the methylated and unmethylated probe intensities in your tab-delimited files. The beta-values were basically calculated from the methylated and unmethylated probe intensities. I recommend using .IDAT files if you have them because it includes all the information. > > As for your question about diffTh parameter, you can think it as the log2-foldChange threshold, which is similar as what expression analysis always do. > > There are couple of reasons why visually check the methylation data (without smoothing) is important. Here is just list a couple of them: > 1. The location of DMRs, because methylation regulation of transcript expression is very much chromosome location dependent. DMRs near TSS and CpG-islands are more likely to be involved in negative regulation of transcript expression. > 2. Check whether there is consistent behavior of nearby probes (if there are) because methylation pattern tends to be locally correlated. > 3. Globally check the methylation difference across samples. This may help the quality assessment of the samples, or identify whether there global-hyper or hypo methylation events. > > > Pan > > > On Tue, Feb 25, 2014 at 9:47 AM, Allegra A. Petti <apetti@genome.wustl.edu> wrote: > Hi Pan and Giovanni, > > I realized I have one more question. I'm currently using methyAnalysis starting from idat files, but I'd also like to be able to start from a tab-delimited matrix of beta values. I couldn't figure out how to work with (and annotate) beta values in any of the R packages that produce MethyLumiM objects. Do you have any procedures for this? > > Thank you again, > Allegra > _________________________________ > Allegra A. Petti, Ph.D. > > The Genome Institute > 4444 Forest Park Ave. > St. Louis, MO 63108 > apetti@genome.wustl.edu > allegra.conbrio@post.harvard.edu > > On Feb 25, 2014, at 11:23 AM, Pan Du <dupan.mail@gmail.com> wrote: > >> Hi Allegra and Giovanni >> >> The parameters of identifySigDMR need to be tuned based on your dataset. The default parameters was designed for the cell lines with clear difference between groups. For tissue samples, these parameters need to be less stringent, especially the diffTh parameter. >> Also, I strongly recommend visually checking your data. You can export the methylation data using export.methyGenoSet, and then visualize it in IGV. >> >> Pan >> >> >> >> On Tue, Feb 25, 2014 at 8:31 AM, Giovanni Calice <giovcalice@gmail.com> wrote: >> Hi Allegra, >> >> I'm having some difficulties in identifying Differentially Methylated Regions on my datasets too, >> both with bumphunter/minfi and with methyAnalysis package. >> >> With regard to bumphunter/minfi there was a clarification also to me >> in the previous massages of the BioC list. >> >> With regard to methyAnalysis I think you performed the smoothMethyData function separately >> It isn't wrong; in fact, if you had not smoothing separately, the next method of the pipeline detectDMR.slideWin >> performs it first and then proceeds with differential analysis. >> >> According to me you could try setting a different p.value.detection.th argument and try to see if you get results. >> >> Instead I get results in identifying Differentially Methylated Regions by methyAnalysis but >> none is significant (p.adjust = 1 or p.adjust > 0.7). >> >> So I'm wondering could the small number of samples (I've 8-12 samples for now) affect >> the significance of results and lead to a p.adjust not so good? >> >> Best, >> Giovanni >> >> >> Laboratory of Preclinical and Translational Research >> IRCCS - CROB Oncology Referral Center of Basilicata >> Rionero in Vulture - Italy >> >> *** Mail from ************************ >> Bioconductor Digest, Vol 132, Issue 26 >> ************************************** >> ************************************** >> Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) >> From: "Allegra Petti [guest]" <guest@bioconductor.org> >> To: bioconductor@r-project.org, apetti@genome.wustl.edu >> Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> >> Subject: [BioC] methyAnalysis for DMR identification >> Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> >> >> >> >> Hello, >> >> I was wondering if anyone here has tried methyAnalysis for identifying Differentially Methylated Regions in Illumina 450k array data. It seems to have wonderful data-visualization options, but it found no differentially-methylated regions in my data (56 450k data sets), which was surprising. I've found very little documentation for this method, and I am wondering if I did something wrong - for example, should I have performed the smoothMethyData function separately, with a different window size? Relevant code snippets and output are provided below. I'd greatly appreciate any thoughts or suggestions. >> >> Thank you very much! >> Allegra >> __________________________________________________________ >> Code: >> >> library(lumi); >> library(methylumi); >> library(wateRmelon); >> library(methyAnalysis); >> library(IlluminaHumanMethylation450k.db); >> library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be installed manually >> barcodes <- readLines("barcodes.txt"); # read barcodes from a text file >> classes <- readLines("IdatClasses.txt"); >> data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) >> *** additional data processing *** >> data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object >> data.g <- MethyLumiM2GenoSet(data.m, lib = "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet >> dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data and find DMRs >> allDMRInfo <- identifySigDMR(dmrResult); >> ___________________________________________________________________ >> Standard Output: >> >> 65 probes were removed because of lack of chromosome location information! >> Smoothing Chromosome chr1 ... >> Smoothing Chromosome chr10 ... >> Smoothing Chromosome chr11 ... >> Smoothing Chromosome chr12 ... >> Smoothing Chromosome chr13 ... >> Smoothing Chromosome chr14 ... >> Smoothing Chromosome chr15 ... >> Smoothing Chromosome chr16 ... >> Smoothing Chromosome chr17 ... >> Smoothing Chromosome chr18 ... >> Smoothing Chromosome chr19 ... >> Smoothing Chromosome chr2 ... >> Smoothing Chromosome chr20 ... >> Smoothing Chromosome chr21 ... >> Smoothing Chromosome chr22 ... >> Smoothing Chromosome chr3 ... >> Smoothing Chromosome chr4 ... >> Smoothing Chromosome chr5 ... >> Smoothing Chromosome chr6 ... >> Smoothing Chromosome chr7 ... >> Smoothing Chromosome chr8 ... >> Smoothing Chromosome chr9 ... >> Smoothing Chromosome chrX ... >> Smoothing Chromosome chrY ... >> [1] "Done with detectDMR.slideWin\n" >> No significant CpG-sites were identified based on current criteria! >> >> >> >> -- output of sessionInfo(): >> >> R version 3.0.2 (2013-09-25) >> Platform: x86_64-unknown-linux-gnu (64-bit) >> >> locale: >> [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C >> [3] LC_TIME=en_US.utf8 LC_COLLATE=C >> [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 >> [7] LC_PAPER=en_US.utf8 LC_NAME=C >> [9] LC_ADDRESS=C LC_TELEPHONE=C >> [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C >> >> attached base packages: >> [1] grid parallel methods stats graphics grDevices utils >> [8] datasets base >> >> other attached packages: >> [1] MASS_7.3-29 >> [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 >> [3] minfi_1.9.11 >> [4] bumphunter_1.2.0 >> [5] locfit_1.5-9.1 >> [6] iterators_1.0.6 >> [7] foreach_1.4.1 >> [8] Biostrings_2.30.1 >> [9] lattice_0.20-24 >> [10] methyAnalysis_1.4.2 >> [11] GenomicRanges_1.14.4 >> [12] XVector_0.2.0 >> [13] IRanges_1.20.6 >> [14] wateRmelon_1.2.2 >> [15] ROC_1.38.0 >> [16] IlluminaHumanMethylation450k.db_2.0.7 >> [17] org.Hs.eg.db_2.10.1 >> [18] RSQLite_0.11.4 >> [19] DBI_0.2-7 >> [20] AnnotationDbi_1.24.0 >> [21] limma_3.18.4 >> [22] methylumi_2.8.0 >> [23] matrixStats_0.8.14 >> [24] ggplot2_0.9.3.1 >> [25] reshape2_1.2.2 >> [26] scales_0.2.3 >> [27] lumi_2.14.1 >> [28] Biobase_2.22.0 >> [29] BiocGenerics_0.8.0 >> >> loaded via a namespace (and not attached): >> [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 >> [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 >> [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 >> [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 >> [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 >> [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 >> [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 >> [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 >> [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 >> [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 >> [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 >> [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 >> [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 >> [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 >> 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 >> [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 >> [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 >> [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 >> [55] xtable_1.7-1 zlibbioc_1.8.0 >> >> ************************************** >> ************************************** >> > > > ____ This email message is a private communication. The information transmitted, including attachments, is intended only for the person or entity to which it is addressed and may contain confidential, privileged, and/or proprietary material. Any review, duplication, retransmission, distribution, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is unauthorized by the sender and is prohibited. If you have received this message in error, please contact the sender immediately by return email and delete the original message from all computer systems. Thank you. > ____ This email message is a private communication. The information transmitted, including attachments, is intended only for the person or entity to which it is addressed and may contain confidential, privileged, and/or proprietary material. Any review, duplication, retransmission, distribution, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is unauthorized by the sender and is prohibited. If you have received this message in error, please contact the sender immediately by return email and delete the original message from all computer systems. Thank you. [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Giovanni, I got results using some arbitrarily chosen lenient parameters. I don't know yet how appropriate these parameters are for your data or mine, but in case you want to try them, I used the following command: allDMRInfo <- identifySigDMR(dmrResult, p.adjust.method="fdr", pValueTh=0.05, fdrTh=0.1, diffTh=0.1) (My data set has 56 samples, but in my uninformed opinion, it doesn't seem like 8-12 samples is too few.) Best, Allegra On 02/25/2014 10:31 AM, Giovanni Calice wrote: > Hi Allegra, > > I'm having some difficulties in identifying Differentially Methylated > Regions on my datasets too, > both with bumphunter/minfi and with methyAnalysis package. > > With regard to bumphunter/minfi there was a clarification also to me > in the previous massages of the BioC list. > > With regard to methyAnalysis I think you performed the smoothMethyData > function separately > It isn't wrong; in fact, if you had not smoothing separately, the next > method of the pipeline detectDMR.slideWin > performs it first and then proceeds with differential analysis. > > According to me you could try setting a different p.value.detection.th > <http: p.value.detection.th=""> argument and try to see if you get results. > > Instead I get results in identifying Differentially Methylated Regions > by methyAnalysis but > none is significant (p.adjust = 1 or p.adjust > 0.7). > > So I'm wondering could the small number of samples (I've 8-12 samples > for now) affect > the significance of results and lead to a p.adjust not so good? > > Best, > Giovanni > > > Laboratory of Preclinical and Translational Research > IRCCS - CROB Oncology Referral Center of Basilicata > Rionero in Vulture - Italy > > *** Mail from ************************ > Bioconductor Digest, Vol 132, Issue 26 > ************************************** > ************************************** > Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) > From: "Allegra Petti [guest]" <guest@bioconductor.org> <mailto:guest@bioconductor.org>> > To: bioconductor@r-project.org <mailto:bioconductor@r-project.org>, > apetti@genome.wustl.edu <mailto:apetti@genome.wustl.edu> > Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> <mailto:dupan.mail@gmail.com>> > Subject: [BioC] methyAnalysis for DMR identification > Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org > <mailto:20140224225258.99bc91468a7@mamba.fhcrc.org>> > > > Hello, > > I was wondering if anyone here has tried methyAnalysis for identifying > Differentially Methylated Regions in Illumina 450k array data. It > seems to have wonderful data-visualization options, but it found no > differentially-methylated regions in my data (56 450k data sets), > which was surprising. I've found very little documentation for this > method, and I am wondering if I did something wrong - for example, > should I have performed the smoothMethyData function separately, with > a different window size? Relevant code snippets and output are > provided below. I'd greatly appreciate any thoughts or suggestions. > > Thank you very much! > Allegra > __________________________________________________________ > Code: > > library(lumi); > library(methylumi); > library(wateRmelon); > library(methyAnalysis); > library(IlluminaHumanMethylation450k.db); > library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to > be installed manually > barcodes <- readLines("barcodes.txt"); # read barcodes from a text file > classes <- readLines("IdatClasses.txt"); > data.s <- methylumIDAT(barcodes); # read in idat files (methylumi > function) > *** additional data processing *** > data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object > data.g <- MethyLumiM2GenoSet(data.m, lib = > "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet > dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth > data and find DMRs > allDMRInfo <- identifySigDMR(dmrResult); > ___________________________________________________________________ > Standard Output: > > 65 probes were removed because of lack of chromosome location information! > Smoothing Chromosome chr1 ... > Smoothing Chromosome chr10 ... > Smoothing Chromosome chr11 ... > Smoothing Chromosome chr12 ... > Smoothing Chromosome chr13 ... > Smoothing Chromosome chr14 ... > Smoothing Chromosome chr15 ... > Smoothing Chromosome chr16 ... > Smoothing Chromosome chr17 ... > Smoothing Chromosome chr18 ... > Smoothing Chromosome chr19 ... > Smoothing Chromosome chr2 ... > Smoothing Chromosome chr20 ... > Smoothing Chromosome chr21 ... > Smoothing Chromosome chr22 ... > Smoothing Chromosome chr3 ... > Smoothing Chromosome chr4 ... > Smoothing Chromosome chr5 ... > Smoothing Chromosome chr6 ... > Smoothing Chromosome chr7 ... > Smoothing Chromosome chr8 ... > Smoothing Chromosome chr9 ... > Smoothing Chromosome chrX ... > Smoothing Chromosome chrY ... > [1] "Done with detectDMR.slideWin\n" > No significant CpG-sites were identified based on current criteria! > > > > -- output of sessionInfo(): > > R version 3.0.2 (2013-09-25) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C > [3] LC_TIME=en_US.utf8 LC_COLLATE=C > [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 > [7] LC_PAPER=en_US.utf8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C > > attached base packages: > [1] grid parallel methods stats graphics grDevices utils > [8] datasets base > > other attached packages: > [1] MASS_7.3-29 > [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 > [3] minfi_1.9.11 > [4] bumphunter_1.2.0 > [5] locfit_1.5-9.1 > [6] iterators_1.0.6 > [7] foreach_1.4.1 > [8] Biostrings_2.30.1 > [9] lattice_0.20-24 > [10] methyAnalysis_1.4.2 > [11] GenomicRanges_1.14.4 > [12] XVector_0.2.0 > [13] IRanges_1.20.6 > [14] wateRmelon_1.2.2 > [15] ROC_1.38.0 > [16] IlluminaHumanMethylation450k.db_2.0.7 > [17] org.Hs.eg.db_2.10.1 > [18] RSQLite_0.11.4 > [19] DBI_0.2-7 > [20] AnnotationDbi_1.24.0 > [21] limma_3.18.4 > [22] methylumi_2.8.0 > [23] matrixStats_0.8.14 > [24] ggplot2_0.9.3.1 > [25] reshape2_1.2.2 > [26] scales_0.2.3 > [27] lumi_2.14.1 > [28] Biobase_2.22.0 > [29] BiocGenerics_0.8.0 > > loaded via a namespace (and not attached): > [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 > [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 > [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 > [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 > [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 > [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 > [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 > [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 > [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 > [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 > [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 > [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 > [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 > [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 > 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 > [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 > [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 > [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 > [55] xtable_1.7-1 zlibbioc_1.8.0 > > ************************************** > ************************************** ____ This email message is a private communication. The information transmitted, including attachments, is intended only for the person or entity to which it is addressed and may contain confidential, privileged, and/or proprietary material. Any review, duplication, retransmission, distribution, or other use of, or taking of any action in reliance upon, this information by persons or entities other than the intended recipient is unauthorized by the sender and is prohibited. If you have received this message in error, please contact the sender immediately by return email and delete the original message from all computer systems. Thank you. [[alternative HTML version deleted]]
ADD REPLY
0
Entering edit mode
Hi Allegra, I have already tried to filter with the outcome of No significant CpG- sites identified. Most likely it is the small number of dataset size that influences the results. I'll follow Pan's advice to visualize-check data in IGV and I'm waiting for other samples of the next Illumina chip-array by my colleagues in such way to execute methyAnalysis pipeline on a dataset with a doubled size (20-24), so finally have significant CpG-sites. Best, Giovanni 2014-02-25 19:11 GMT+01:00 Allegra Petti <apetti@genome.wustl.edu>: > Hi Giovanni, > > I got results using some arbitrarily chosen lenient parameters. I don't > know yet how appropriate these parameters are for your data or mine, but in > case you want to try them, I used the following command: > > allDMRInfo <- identifySigDMR(dmrResult, p.adjust.method="fdr", > pValueTh=0.05, fdrTh=0.1, diffTh=0.1) > > (My data set has 56 samples, but in my uninformed opinion, it doesn't seem > like 8-12 samples is too few.) > > Best, > Allegra > > > On 02/25/2014 10:31 AM, Giovanni Calice wrote: > > Hi Allegra, > > I'm having some difficulties in identifying Differentially Methylated > Regions on my datasets too, > both with bumphunter/minfi and with methyAnalysis package. > > With regard to bumphunter/minfi there was a clarification also to me > in the previous massages of the BioC list. > > With regard to methyAnalysis I think you performed the smoothMethyData > function separately > It isn't wrong; in fact, if you had not smoothing separately, the next > method of the pipeline detectDMR.slideWin > performs it first and then proceeds with differential analysis. > > According to me you could try setting a different p.value.detection.thargument and try to see if you get results. > > Instead I get results in identifying Differentially Methylated Regions by > methyAnalysis but > none is significant (p.adjust = 1 or p.adjust > 0.7). > > So I'm wondering could the small number of samples (I've 8-12 samples for > now) affect > the significance of results and lead to a p.adjust not so good? > > Best, > Giovanni > > > Laboratory of Preclinical and Translational Research > IRCCS - CROB Oncology Referral Center of Basilicata > Rionero in Vulture - Italy > > *** Mail from ************************ > Bioconductor Digest, Vol 132, Issue 26 > ************************************** > ************************************** > Date: Mon, 24 Feb 2014 14:52:58 -0800 (PST) > From: "Allegra Petti [guest]" <guest@bioconductor.org> > To: bioconductor@r-project.org, apetti@genome.wustl.edu > Cc: methyAnalysis Maintainer <dupan.mail@gmail.com> > Subject: [BioC] methyAnalysis for DMR identification > Message-ID: <20140224225258.99BC91468A7@mamba.fhcrc.org> > > > Hello, > > I was wondering if anyone here has tried methyAnalysis for identifying > Differentially Methylated Regions in Illumina 450k array data. It seems to > have wonderful data-visualization options, but it found no > differentially-methylated regions in my data (56 450k data sets), which was > surprising. I've found very little documentation for this method, and I am > wondering if I did something wrong - for example, should I have performed > the smoothMethyData function separately, with a different window size? > Relevant code snippets and output are provided below. I'd greatly > appreciate any thoughts or suggestions. > > Thank you very much! > Allegra > __________________________________________________________ > Code: > > library(lumi); > library(methylumi); > library(wateRmelon); > library(methyAnalysis); > library(IlluminaHumanMethylation450k.db); > library(IlluminaHumanMethylation450kanno.ilmn12.hg19); # Note: had to be > installed manually > barcodes <- readLines("barcodes.txt"); # read barcodes from a text file > classes <- readLines("IdatClasses.txt"); > data.s <- methylumIDAT(barcodes); # read in idat files (methylumi function) > *** additional data processing *** > data.m <- as(data.s, "MethyLumiM"); # convert to MethyLumiM object > data.g <- MethyLumiM2GenoSet(data.m, lib = > "IlluminaHumanMethylation450k.db"); # convert MethyLumiM to MethyGenoSet > dmrResult <- detectDMR.slideWin(data.g, sampleType=classes); # smooth data > and find DMRs > allDMRInfo <- identifySigDMR(dmrResult); > ___________________________________________________________________ > Standard Output: > > 65 probes were removed because of lack of chromosome location information! > Smoothing Chromosome chr1 ... > Smoothing Chromosome chr10 ... > Smoothing Chromosome chr11 ... > Smoothing Chromosome chr12 ... > Smoothing Chromosome chr13 ... > Smoothing Chromosome chr14 ... > Smoothing Chromosome chr15 ... > Smoothing Chromosome chr16 ... > Smoothing Chromosome chr17 ... > Smoothing Chromosome chr18 ... > Smoothing Chromosome chr19 ... > Smoothing Chromosome chr2 ... > Smoothing Chromosome chr20 ... > Smoothing Chromosome chr21 ... > Smoothing Chromosome chr22 ... > Smoothing Chromosome chr3 ... > Smoothing Chromosome chr4 ... > Smoothing Chromosome chr5 ... > Smoothing Chromosome chr6 ... > Smoothing Chromosome chr7 ... > Smoothing Chromosome chr8 ... > Smoothing Chromosome chr9 ... > Smoothing Chromosome chrX ... > Smoothing Chromosome chrY ... > [1] "Done with detectDMR.slideWin\n" > No significant CpG-sites were identified based on current criteria! > > > > -- output of sessionInfo(): > > R version 3.0.2 (2013-09-25) > Platform: x86_64-unknown-linux-gnu (64-bit) > > locale: > [1] LC_CTYPE=en_US.utf8 LC_NUMERIC=C > [3] LC_TIME=en_US.utf8 LC_COLLATE=C > [5] LC_MONETARY=en_US.utf8 LC_MESSAGES=en_US.utf8 > [7] LC_PAPER=en_US.utf8 LC_NAME=C > [9] LC_ADDRESS=C LC_TELEPHONE=C > [11] LC_MEASUREMENT=en_US.utf8 LC_IDENTIFICATION=C > > attached base packages: > [1] grid parallel methods stats graphics grDevices utils > [8] datasets base > > other attached packages: > [1] MASS_7.3-29 > [2] IlluminaHumanMethylation450kanno.ilmn12.hg19_0.2.1 > [3] minfi_1.9.11 > [4] bumphunter_1.2.0 > [5] locfit_1.5-9.1 > [6] iterators_1.0.6 > [7] foreach_1.4.1 > [8] Biostrings_2.30.1 > [9] lattice_0.20-24 > [10] methyAnalysis_1.4.2 > [11] GenomicRanges_1.14.4 > [12] XVector_0.2.0 > [13] IRanges_1.20.6 > [14] wateRmelon_1.2.2 > [15] ROC_1.38.0 > [16] IlluminaHumanMethylation450k.db_2.0.7 > [17] org.Hs.eg.db_2.10.1 > [18] RSQLite_0.11.4 > [19] DBI_0.2-7 > [20] AnnotationDbi_1.24.0 > [21] limma_3.18.4 > [22] methylumi_2.8.0 > [23] matrixStats_0.8.14 > [24] ggplot2_0.9.3.1 > [25] reshape2_1.2.2 > [26] scales_0.2.3 > [27] lumi_2.14.1 > [28] Biobase_2.22.0 > [29] BiocGenerics_0.8.0 > > loaded via a namespace (and not attached): > [1] BSgenome_1.30.0 BiocInstaller_1.12.0 Formula_1.1-1 > [4] GenomicFeatures_1.14.2 Gviz_1.6.0 Hmisc_3.14-0 > [7] KernSmooth_2.23-10 Matrix_1.1-2 R.methodsS3_1.6.1 > [10] RColorBrewer_1.0-5 RCurl_1.95-4.1 Rsamtools_1.14.2 > [13] XML_3.98-1.1 affy_1.40.0 affyio_1.30.0 > [16] annotate_1.40.0 beanplot_1.1 biomaRt_2.18.0 > [19] biovizBase_1.10.7 bitops_1.0-6 cluster_1.14.4 > [22] codetools_0.2-8 colorspace_1.2-4 dichromat_2.0-0 > [25] digest_0.6.4 doRNG_1.5.5 genefilter_1.44.0 > [28] genoset_1.14.0 gtable_0.1.2 illuminaio_0.2.0 > [31] itertools_0.1-1 labeling_0.2 latticeExtra_0.6-26 > [34] mclust_4.2 mgcv_1.7-28 multtest_2.18.0 > [37] munsell_0.4.2 nleqslv_2.1 nlme_3.1-113 > [40] nor1mix_1.1-4 pkgmaker_0.17.4 plyr_1.8 > 43] preprocessCore_1.24.0 proto_0.3-10 registry_0.2 > [46] reshape_0.8.4 rngtools_1.2.3 rtracklayer_1.22.3 > [49] siggenes_1.36.0 splines_3.0.2 stats4_3.0.2 > [52] stringr_0.6.2 survival_2.37-7 tools_3.0.2 > [55] xtable_1.7-1 zlibbioc_1.8.0 > > ************************************** > ************************************** > > > > ____ This email message is a private communication. The information > transmitted, including attachments, is intended only for the person or > entity to which it is addressed and may contain confidential, privileged, > and/or proprietary material. Any review, duplication, retransmission, > distribution, or other use of, or taking of any action in reliance upon, > this information by persons or entities other than the intended recipient > is unauthorized by the sender and is prohibited. If you have received this > message in error, please contact the sender immediately by return email and > delete the original message from all computer systems. Thank you. > [[alternative HTML version deleted]]
ADD REPLY

Login before adding your answer.

Traffic: 740 users visited in the last hour
Help About
FAQ
Access RSS
API
Stats

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

Powered by the version 2.3.6