vignettes/DecompPipeline.md
In lutsik/DecompPipeline: Preparation pipeline for MeDeCom
title: 'DecompPipeline: Preprocessing of DNA Methylation data for MeDeCom'
author: "Michael Scherer, Pavlo Lutsik"
date: '2019-09-27'
output:
html_document:
fig_height: 5
fig_width: 5
keep_md: yes
mathjax: default
number_sections: no
toc: yes
pdf_document:
toc: yes
bibliography: biblio.bib
vignette: >
%\VignetteIndexEntry{MeDeCom}
%\VignetteEngine{knitr::rmarkdown}
\usepackage[utf8]{inputenc}
Introduction
DecompPipeline is an R package created for preprocessing DNA Methylation data for deconvolution of complex tissue samples using one of the methods MeDeCom, RefFreeEWAS or EDec. Briefly, non-negative matrix factorization of complex methylation data sets is performed to discover latent methylation components (LMCs). Those components can, for instance, represent cell types, but are not limited to that. The DecompPipeline uses the RnBeads package for handling input DNA methylation data and is able to handle both BeadArray (27k, 450k, EPIC) and bisulfite sequencing (RRBS, WGBS) data. Necessary steps include stringent filtering and selecting the correct subset of CpG sites for downstream MeDeCom analysis. Also check out FactorViz for the visualization of deconvolution results.
Installation
You can install the DecompPipeline through GitHub using devtools:
install.packages("devtools",repos="https://cran.uni-muenster.de/")
devtools::install_github("lutsik/DecompPipeline")
Using DecompPipeline
Default Pipeline
The main workhorse for DecompPipeline is start_decomp_pipeline. It only requires few inputs, including DNA methylation data either in the form of a matrix\data.frame or as an RnBSet object. We will now discuss the rich functionalities that are available for having and RnBSet as input. A short introduction on how to import DNA methylation data using RnBeads is given in the last section. For details on the many options available within the DecompPipeline, see the function's documentation. We will further explain the options in this vignette.
CpG Filtering
This filtering step involves removing potentially unreliable and/or problematic CpGs from further analysis and has dedicated functions both for bisulfite sequencing and array based data sets. We will discuss the two data types separately:
Array based data sets
Filtering CpG sites of array based data sets (27k, 450k, EPIC) involves setting the minimum number of required beads on the chip (MIN_N_BEADS). Furthermore, low and high intensity outliers can be removed by a quantile approach, which removes the highest (MAX_INT_QUANT) and lowest quantile (MIN_INT_QUANT). In addition, all sites containing any missing value (FILTER_NA), outside of CpG context (FILTER_CONTEXT), mapping to an annotated Single Nucleotide Polymorphism (FILTER_SNP, snp.list) and on the sex chromosomes (FILTER_SOMATIC) can be omitted. Further options are available and described in the function's documentation. The function also provides options to normalize data using the methods available in the RnBeads R package and will return the processed data set and further information on the steps executed.
data("small.RnBeadSet")
data.prep <- prepare_data(RNB_SET = rnb.set.example,
NORMALIZATION = "wm.dasen",
MIN_N_BEADS = 5,
MIN_INT_QUANT = 0.05,
MAX_INT_QUANT = 0.95,
FILTER_NA = T,
FILTER_SNP = T,
FILTER_CONTEXT = FALSE,
FILTER_SOMATIC = FALSE)
names(data.prep)
## [1] "quality.filter" "annot.filter" "total.filter"
## [4] "rnb.set.filtered" "info"
Bisulfite sequencing based data sets
For bisulfite sequencing data sets, different filtering criteria apply. First, a absolute coverage threshold can be specified with MIN_COVERAGE to remove all sites with lower coverage. Similar to array-based data sets, upper and lower quantile of coverage can be omitted using MIN_COVG_QUANT and MAX_COVG_QUANT. In complete accordance with array-based data sets, sites having missing values, located at annotated SNPs and on sex chromosomes can be removed.
rnb.set <- load.rnb.set(system.file("extdata/small_rnbSet.zip",package="DecompPipeline"))
data.prep.bs <- prepare_data_BS(RNB_SET = rnb.set,
MIN_COVERAGE = 5,
MIN_COVG_QUANT = 0.1,
MAX_COVG_QUANT = 0.9,
FILTER_NA = T,
FILTER_SOMATIC = F,
FILTER_SNP = F)
## opening ff /tmp/Rtmpw8TrQZ/ff6f623a453210.ff
## opening ff /tmp/Rtmpw8TrQZ/ff6f627581c0dd.ff
names(data.prep.bs)
## [1] "quality.filter" "annot.filter" "total.filter"
## [4] "rnb.set.filtered"
Selecting subsets of CpGs
Since performing MeDeCom on complete 450k/EPIC or BS datasets is still computationally infeasible, it is crucial to select sites for subsequent analysis that might define LMCs. The DecompPipeline provides multiple options to preselect those CpGs and we will briefly introduce each of them:
- pheno This option selects the markers that define sample identity for the grouping information given through rnb.sample.groups. Briefly, the limma method is used to define differentially methylated sites between the two groups and uses those markers for subsequent analysis.
- houseman2012 This option selects 50,000 sites that were found to be cell-type specific using the @houseman_refbased method on blood cell types. The Houseman method was employed on the @reiniusRef dataset and is thus only applicable to blood data sets.
- houseman2014 Here, sites are selected as cell-type specific by RefFreeEWAS. For further information, see @Houseman2014. This method is applicable to any kind of data set.
- jaffe2014 Another list of supposedly cell-type specific CpG sites reported in @Jaffe2014.
- rowFstat If reference methylation profiles are provided through
REF_DATA_SET and REF_PHENO_COLUMN, sites are selected as those being linked to the reference cell types using an F-test.
- random Sites are randomly selected from all possible sites.
- pca This option selects the sites that have most influence on the principal components. The number of principal components calculated is determined by
N_PRIN_COMP.
- var The most variable sites across all samples are selected and used for subsequent analysis (DEFAULT option).
- hybrid Selects half of the sites randomly and the other half as the most variable.
- range This options selects the sites that have a difference between the minimum and maximum value across all samples higher than
RANGE_DIFF.
- custom The sites to be used are provided by the user with a file containing row indices. The file needs to be provided in
CUSTOM_MARKER_FILE.
- all Using all sites available in the input.
- pcadapt Uses principal component analysis as implemented in the bigstats R package to determine sites that are significantly linked to the potential cell types. This requires specifying K a priori (argument
K.prior). We thank Florian Prive and Sophie Achard for providing the idea and parts of the codes.
- edec_stage0 Employs EDec's stage 0 to infer cell-type specific markers. By default EDec's example reference data is provided.
For most of the options (except for houseman2012, jaffe2014, and range) the number of selected sites can be specified using the parameter N_MARKERS. In contrast to CpG filtering, subset selection is independent of the data type (array-based and BS). The function returns a list, with each entry containing row indices of the selected sites:
cg_subsets <- prepare_CG_subsets(rnb.set=data.prep$rnb.set.filtered,
MARKER_SELECTION = c("houseman2012","var"),
N_MARKERS = 500
)
lengths(cg_subsets)
## houseman2012 var
## 110 500
Starting MeDeCom
After these preprocessing steps, you are ready to perfom the actual MeDeCom analysis using the start_medecom_analysis function. To store output in a format that is later on readable by FactorViz, you need to set the flag factorviz.outputs. Further parameters are described in detail in the reference manual.
md.res <- start_medecom_analysis(rnb.set=data.prep$rnb.set.filtered,
cg_groups = cg_subsets,
Ks=2:5,
LAMBDA_GRID = c(0.01,0.001),
factorviz.outputs = T)
## [1] "Did not write the variable dump: should only be executed from an environment with all the variables set"
## [2019-09-27 14:42:04, Main:] checking inputs
## [2019-09-27 14:42:04, Main:] preparing data
## [2019-09-27 14:42:04, Main:] preparing jobs
## [2019-09-27 14:42:04, Main:] 176 factorization runs in total
## [2019-09-27 14:48:59, Main:] 100 runs complete
## [2019-09-27 15:01:21, Main:] finished all jobs. Creating the object
Executing DecompPipeline
You can also peform all the steps above, by just calling a single function:
md.res <- start_decomp_pipeline(rnb.set=rnb.set,
Ks=2:5,
lambda.grid = c(0.01,0.001),
factorviz.outputs = T,
marker.selection = c("houseman2012","var"),
n.markers = 50,
min.n.beads = 5,
min.int.quant = 0.05,
max.int.quant = 0.95,
filter.na = T,
filter.snp = T,
filter.context = FALSE,
filter.somatic = FALSE,
normalization="wm.dasen")
## [1] "Did not write the variable dump: should only be executed from an environment with all the variables set"
## [2019-09-27 15:02:03, Main:] checking inputs
## [2019-09-27 15:02:03, Main:] preparing data
## [2019-09-27 15:02:03, Main:] preparing jobs
## [2019-09-27 15:02:03, Main:] 176 factorization runs in total
## [2019-09-27 15:02:58, Main:] 100 runs complete
## [2019-09-27 15:03:30, Main:] finished all jobs. Creating the object
Data Import through RnBeads
We recommend to use the RnBeads package to provide methylation data to DecompPipeline. RnBeads can handle BS and array-based datasets and provides an extensive toolset. BS data can be loaded directly from BED-files generated through methylation data mapping and calling software, such as bismark or bsmap. For array-based datasets, IDAT-files can be directly loaded or GEO accession numbers provided to download data from the repository, among other import options. We refer to the RnBeads vignette for further descriptions on the data import options. Data Import is the only module to be exectued ahead of using DecompPipeline.
idat.dir <- "~/idats"
sample.annotation <- "~/sample_annotation.csv"
rnb.set <- rnb.execute.import(data.source = c(idat.dir,sample.annotation),data.type = "infinium.idat.dir")
R session
Here is the output of sessionInfo() on the system on which this document was compiled:
## R version 3.5.3 (2019-03-11)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Debian GNU/Linux 8 (jessie)
##
## Matrix products: default
## BLAS: /usr/lib/atlas-base/atlas/libblas.so.3.0
## LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] grid stats4 parallel stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] RnBeads.hg19_1.14.0
## [2] DecompPipeline_0.2.5
## [3] R.utils_2.7.0
## [4] R.oo_1.22.0
## [5] R.methodsS3_1.7.1
## [6] MeDeCom_0.2.2
## [7] RnBeads_2.3.1
## [8] plyr_1.8.4
## [9] methylumi_2.28.0
## [10] minfi_1.28.0
## [11] bumphunter_1.24.5
## [12] locfit_1.5-9.1
## [13] iterators_1.0.10
## [14] foreach_1.4.4
## [15] Biostrings_2.50.1
## [16] XVector_0.22.0
## [17] SummarizedExperiment_1.12.0
## [18] DelayedArray_0.8.0
## [19] BiocParallel_1.16.2
## [20] FDb.InfiniumMethylation.hg19_2.2.0
## [21] org.Hs.eg.db_3.7.0
## [22] TxDb.Hsapiens.UCSC.hg19.knownGene_3.2.2
## [23] GenomicFeatures_1.34.1
## [24] AnnotationDbi_1.44.0
## [25] reshape2_1.4.3
## [26] scales_1.0.0
## [27] Biobase_2.42.0
## [28] illuminaio_0.24.0
## [29] matrixStats_0.54.0
## [30] limma_3.38.2
## [31] gridExtra_2.3
## [32] ggplot2_3.1.0
## [33] fields_9.6
## [34] maps_3.3.0
## [35] spam_2.2-0
## [36] dotCall64_1.0-0
## [37] ff_2.2-14
## [38] bit_1.1-14
## [39] cluster_2.0.7-1
## [40] MASS_7.3-51.1
## [41] GenomicRanges_1.34.0
## [42] GenomeInfoDb_1.18.1
## [43] IRanges_2.16.0
## [44] S4Vectors_0.20.1
## [45] BiocGenerics_0.28.0
## [46] RUnit_0.4.32
## [47] gplots_3.0.1
## [48] gtools_3.8.1
## [49] pracma_2.2.2
## [50] Rcpp_1.0.0
## [51] usethis_1.4.0
## [52] devtools_2.0.1
## [53] knitr_1.21
##
## loaded via a namespace (and not attached):
## [1] backports_1.1.2 lazyeval_0.2.1
## [3] splines_3.5.3 digest_0.6.18
## [5] gdata_2.18.0 magrittr_1.5
## [7] memoise_1.1.0 remotes_2.0.2
## [9] readr_1.2.1 annotate_1.60.0
## [11] siggenes_1.56.0 prettyunits_1.0.2
## [13] colorspace_1.3-2 blob_1.1.1
## [15] xfun_0.4 dplyr_0.7.8
## [17] callr_3.0.0 crayon_1.3.4
## [19] RCurl_1.95-4.11 genefilter_1.64.0
## [21] bindr_0.1.1 GEOquery_2.50.0
## [23] survival_2.43-3 glue_1.3.0
## [25] registry_0.5 gtable_0.2.0
## [27] zlibbioc_1.28.0 pkgbuild_1.0.2
## [29] Rhdf5lib_1.4.1 HDF5Array_1.10.0
## [31] DBI_1.0.0 rngtools_1.3.1
## [33] bibtex_0.4.2 xtable_1.8-3
## [35] progress_1.2.0 mclust_5.4.2
## [37] preprocessCore_1.44.0 httr_1.3.1
## [39] RColorBrewer_1.1-2 pkgconfig_2.0.2
## [41] reshape_0.8.8 XML_3.98-1.16
## [43] tidyselect_0.2.5 rlang_0.3.1
## [45] munsell_0.5.0 tools_3.5.3
## [47] cli_1.0.1 RSQLite_2.1.1
## [49] evaluate_0.12 stringr_1.3.1
## [51] processx_3.2.0 bit64_0.9-7
## [53] fs_1.2.6 beanplot_1.2
## [55] caTools_1.17.1.1 purrr_0.2.5
## [57] bindrcpp_0.2.2 nlme_3.1-137
## [59] doRNG_1.7.1 nor1mix_1.2-3
## [61] xml2_1.2.0 biomaRt_2.38.0
## [63] compiler_3.5.3 tibble_1.4.2
## [65] stringi_1.2.4 ps_1.2.1
## [67] desc_1.2.0 lattice_0.20-38
## [69] Matrix_1.2-16 multtest_2.38.0
## [71] pillar_1.3.0 data.table_1.11.8
## [73] bitops_1.0-6 rtracklayer_1.42.1
## [75] R6_2.3.0 KernSmooth_2.23-15
## [77] sessioninfo_1.1.1 codetools_0.2-15
## [79] assertthat_0.2.0 pkgload_1.0.2
## [81] rhdf5_2.26.0 openssl_1.1
## [83] pkgmaker_0.27 rprojroot_1.3-2
## [85] withr_2.1.2 GenomicAlignments_1.18.0
## [87] Rsamtools_1.34.0 GenomeInfoDbData_1.2.0
## [89] hms_0.4.2 quadprog_1.5-5
## [91] tidyr_0.8.2 base64_2.0
## [93] DelayedMatrixStats_1.4.0 base64enc_0.1-3
References
lutsik/DecompPipeline documentation built on Oct. 13, 2019, 1:51 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
title: 'DecompPipeline: Preprocessing of DNA Methylation data for MeDeCom' author: "Michael Scherer, Pavlo Lutsik" date: '2019-09-27' output: html_document: fig_height: 5 fig_width: 5 keep_md: yes mathjax: default number_sections: no toc: yes pdf_document: toc: yes bibliography: biblio.bib vignette: > %\VignetteIndexEntry{MeDeCom} %\VignetteEngine{knitr::rmarkdown} \usepackage[utf8]{inputenc}
Introduction
DecompPipeline is an R package created for preprocessing DNA Methylation data for deconvolution of complex tissue samples using one of the methods MeDeCom, RefFreeEWAS or EDec. Briefly, non-negative matrix factorization of complex methylation data sets is performed to discover latent methylation components (LMCs). Those components can, for instance, represent cell types, but are not limited to that. The DecompPipeline uses the RnBeads package for handling input DNA methylation data and is able to handle both BeadArray (27k, 450k, EPIC) and bisulfite sequencing (RRBS, WGBS) data. Necessary steps include stringent filtering and selecting the correct subset of CpG sites for downstream MeDeCom analysis. Also check out FactorViz for the visualization of deconvolution results.
Installation
You can install the DecompPipeline through GitHub using devtools:
install.packages("devtools",repos="https://cran.uni-muenster.de/")
devtools::install_github("lutsik/DecompPipeline")
Using DecompPipeline
Default Pipeline
The main workhorse for DecompPipeline is start_decomp_pipeline. It only requires few inputs, including DNA methylation data either in the form of a matrix\data.frame or as an RnBSet object. We will now discuss the rich functionalities that are available for having and RnBSet as input. A short introduction on how to import DNA methylation data using RnBeads is given in the last section. For details on the many options available within the DecompPipeline, see the function's documentation. We will further explain the options in this vignette.
CpG Filtering
This filtering step involves removing potentially unreliable and/or problematic CpGs from further analysis and has dedicated functions both for bisulfite sequencing and array based data sets. We will discuss the two data types separately:
Array based data sets
Filtering CpG sites of array based data sets (27k, 450k, EPIC) involves setting the minimum number of required beads on the chip (MIN_N_BEADS). Furthermore, low and high intensity outliers can be removed by a quantile approach, which removes the highest (MAX_INT_QUANT) and lowest quantile (MIN_INT_QUANT). In addition, all sites containing any missing value (FILTER_NA), outside of CpG context (FILTER_CONTEXT), mapping to an annotated Single Nucleotide Polymorphism (FILTER_SNP, snp.list) and on the sex chromosomes (FILTER_SOMATIC) can be omitted. Further options are available and described in the function's documentation. The function also provides options to normalize data using the methods available in the RnBeads R package and will return the processed data set and further information on the steps executed.
data("small.RnBeadSet")
data.prep <- prepare_data(RNB_SET = rnb.set.example,
NORMALIZATION = "wm.dasen",
MIN_N_BEADS = 5,
MIN_INT_QUANT = 0.05,
MAX_INT_QUANT = 0.95,
FILTER_NA = T,
FILTER_SNP = T,
FILTER_CONTEXT = FALSE,
FILTER_SOMATIC = FALSE)
names(data.prep)
## [1] "quality.filter" "annot.filter" "total.filter"
## [4] "rnb.set.filtered" "info"
Bisulfite sequencing based data sets
For bisulfite sequencing data sets, different filtering criteria apply. First, a absolute coverage threshold can be specified with MIN_COVERAGE to remove all sites with lower coverage. Similar to array-based data sets, upper and lower quantile of coverage can be omitted using MIN_COVG_QUANT and MAX_COVG_QUANT. In complete accordance with array-based data sets, sites having missing values, located at annotated SNPs and on sex chromosomes can be removed.
rnb.set <- load.rnb.set(system.file("extdata/small_rnbSet.zip",package="DecompPipeline"))
data.prep.bs <- prepare_data_BS(RNB_SET = rnb.set,
MIN_COVERAGE = 5,
MIN_COVG_QUANT = 0.1,
MAX_COVG_QUANT = 0.9,
FILTER_NA = T,
FILTER_SOMATIC = F,
FILTER_SNP = F)
## opening ff /tmp/Rtmpw8TrQZ/ff6f623a453210.ff
## opening ff /tmp/Rtmpw8TrQZ/ff6f627581c0dd.ff
names(data.prep.bs)
## [1] "quality.filter" "annot.filter" "total.filter"
## [4] "rnb.set.filtered"
Selecting subsets of CpGs
Since performing MeDeCom on complete 450k/EPIC or BS datasets is still computationally infeasible, it is crucial to select sites for subsequent analysis that might define LMCs. The DecompPipeline provides multiple options to preselect those CpGs and we will briefly introduce each of them:
- pheno This option selects the markers that define sample identity for the grouping information given through rnb.sample.groups. Briefly, the limma method is used to define differentially methylated sites between the two groups and uses those markers for subsequent analysis.
- houseman2012 This option selects 50,000 sites that were found to be cell-type specific using the @houseman_refbased method on blood cell types. The Houseman method was employed on the @reiniusRef dataset and is thus only applicable to blood data sets.
- houseman2014 Here, sites are selected as cell-type specific by RefFreeEWAS. For further information, see @Houseman2014. This method is applicable to any kind of data set.
- jaffe2014 Another list of supposedly cell-type specific CpG sites reported in @Jaffe2014.
- rowFstat If reference methylation profiles are provided through
REF_DATA_SETandREF_PHENO_COLUMN, sites are selected as those being linked to the reference cell types using an F-test. - random Sites are randomly selected from all possible sites.
- pca This option selects the sites that have most influence on the principal components. The number of principal components calculated is determined by
N_PRIN_COMP. - var The most variable sites across all samples are selected and used for subsequent analysis (DEFAULT option).
- hybrid Selects half of the sites randomly and the other half as the most variable.
- range This options selects the sites that have a difference between the minimum and maximum value across all samples higher than
RANGE_DIFF. - custom The sites to be used are provided by the user with a file containing row indices. The file needs to be provided in
CUSTOM_MARKER_FILE. - all Using all sites available in the input.
- pcadapt Uses principal component analysis as implemented in the bigstats R package to determine sites that are significantly linked to the potential cell types. This requires specifying K a priori (argument
K.prior). We thank Florian Prive and Sophie Achard for providing the idea and parts of the codes. - edec_stage0 Employs EDec's stage 0 to infer cell-type specific markers. By default EDec's example reference data is provided.
For most of the options (except for houseman2012, jaffe2014, and range) the number of selected sites can be specified using the parameter N_MARKERS. In contrast to CpG filtering, subset selection is independent of the data type (array-based and BS). The function returns a list, with each entry containing row indices of the selected sites:
cg_subsets <- prepare_CG_subsets(rnb.set=data.prep$rnb.set.filtered,
MARKER_SELECTION = c("houseman2012","var"),
N_MARKERS = 500
)
lengths(cg_subsets)
## houseman2012 var
## 110 500
Starting MeDeCom
After these preprocessing steps, you are ready to perfom the actual MeDeCom analysis using the start_medecom_analysis function. To store output in a format that is later on readable by FactorViz, you need to set the flag factorviz.outputs. Further parameters are described in detail in the reference manual.
md.res <- start_medecom_analysis(rnb.set=data.prep$rnb.set.filtered,
cg_groups = cg_subsets,
Ks=2:5,
LAMBDA_GRID = c(0.01,0.001),
factorviz.outputs = T)
## [1] "Did not write the variable dump: should only be executed from an environment with all the variables set"
## [2019-09-27 14:42:04, Main:] checking inputs
## [2019-09-27 14:42:04, Main:] preparing data
## [2019-09-27 14:42:04, Main:] preparing jobs
## [2019-09-27 14:42:04, Main:] 176 factorization runs in total
## [2019-09-27 14:48:59, Main:] 100 runs complete
## [2019-09-27 15:01:21, Main:] finished all jobs. Creating the object
Executing DecompPipeline
You can also peform all the steps above, by just calling a single function:
md.res <- start_decomp_pipeline(rnb.set=rnb.set,
Ks=2:5,
lambda.grid = c(0.01,0.001),
factorviz.outputs = T,
marker.selection = c("houseman2012","var"),
n.markers = 50,
min.n.beads = 5,
min.int.quant = 0.05,
max.int.quant = 0.95,
filter.na = T,
filter.snp = T,
filter.context = FALSE,
filter.somatic = FALSE,
normalization="wm.dasen")
## [1] "Did not write the variable dump: should only be executed from an environment with all the variables set"
## [2019-09-27 15:02:03, Main:] checking inputs
## [2019-09-27 15:02:03, Main:] preparing data
## [2019-09-27 15:02:03, Main:] preparing jobs
## [2019-09-27 15:02:03, Main:] 176 factorization runs in total
## [2019-09-27 15:02:58, Main:] 100 runs complete
## [2019-09-27 15:03:30, Main:] finished all jobs. Creating the object
Data Import through RnBeads
We recommend to use the RnBeads package to provide methylation data to DecompPipeline. RnBeads can handle BS and array-based datasets and provides an extensive toolset. BS data can be loaded directly from BED-files generated through methylation data mapping and calling software, such as bismark or bsmap. For array-based datasets, IDAT-files can be directly loaded or GEO accession numbers provided to download data from the repository, among other import options. We refer to the RnBeads vignette for further descriptions on the data import options. Data Import is the only module to be exectued ahead of using DecompPipeline.
idat.dir <- "~/idats"
sample.annotation <- "~/sample_annotation.csv"
rnb.set <- rnb.execute.import(data.source = c(idat.dir,sample.annotation),data.type = "infinium.idat.dir")
R session
Here is the output of sessionInfo() on the system on which this document was compiled:
## R version 3.5.3 (2019-03-11)
## Platform: x86_64-pc-linux-gnu (64-bit)
## Running under: Debian GNU/Linux 8 (jessie)
##
## Matrix products: default
## BLAS: /usr/lib/atlas-base/atlas/libblas.so.3.0
## LAPACK: /usr/lib/atlas-base/atlas/liblapack.so.3.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=en_US.UTF-8
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## attached base packages:
## [1] grid stats4 parallel stats graphics grDevices utils
## [8] datasets methods base
##
## other attached packages:
## [1] RnBeads.hg19_1.14.0
## [2] DecompPipeline_0.2.5
## [3] R.utils_2.7.0
## [4] R.oo_1.22.0
## [5] R.methodsS3_1.7.1
## [6] MeDeCom_0.2.2
## [7] RnBeads_2.3.1
## [8] plyr_1.8.4
## [9] methylumi_2.28.0
## [10] minfi_1.28.0
## [11] bumphunter_1.24.5
## [12] locfit_1.5-9.1
## [13] iterators_1.0.10
## [14] foreach_1.4.4
## [15] Biostrings_2.50.1
## [16] XVector_0.22.0
## [17] SummarizedExperiment_1.12.0
## [18] DelayedArray_0.8.0
## [19] BiocParallel_1.16.2
## [20] FDb.InfiniumMethylation.hg19_2.2.0
## [21] org.Hs.eg.db_3.7.0
## [22] TxDb.Hsapiens.UCSC.hg19.knownGene_3.2.2
## [23] GenomicFeatures_1.34.1
## [24] AnnotationDbi_1.44.0
## [25] reshape2_1.4.3
## [26] scales_1.0.0
## [27] Biobase_2.42.0
## [28] illuminaio_0.24.0
## [29] matrixStats_0.54.0
## [30] limma_3.38.2
## [31] gridExtra_2.3
## [32] ggplot2_3.1.0
## [33] fields_9.6
## [34] maps_3.3.0
## [35] spam_2.2-0
## [36] dotCall64_1.0-0
## [37] ff_2.2-14
## [38] bit_1.1-14
## [39] cluster_2.0.7-1
## [40] MASS_7.3-51.1
## [41] GenomicRanges_1.34.0
## [42] GenomeInfoDb_1.18.1
## [43] IRanges_2.16.0
## [44] S4Vectors_0.20.1
## [45] BiocGenerics_0.28.0
## [46] RUnit_0.4.32
## [47] gplots_3.0.1
## [48] gtools_3.8.1
## [49] pracma_2.2.2
## [50] Rcpp_1.0.0
## [51] usethis_1.4.0
## [52] devtools_2.0.1
## [53] knitr_1.21
##
## loaded via a namespace (and not attached):
## [1] backports_1.1.2 lazyeval_0.2.1
## [3] splines_3.5.3 digest_0.6.18
## [5] gdata_2.18.0 magrittr_1.5
## [7] memoise_1.1.0 remotes_2.0.2
## [9] readr_1.2.1 annotate_1.60.0
## [11] siggenes_1.56.0 prettyunits_1.0.2
## [13] colorspace_1.3-2 blob_1.1.1
## [15] xfun_0.4 dplyr_0.7.8
## [17] callr_3.0.0 crayon_1.3.4
## [19] RCurl_1.95-4.11 genefilter_1.64.0
## [21] bindr_0.1.1 GEOquery_2.50.0
## [23] survival_2.43-3 glue_1.3.0
## [25] registry_0.5 gtable_0.2.0
## [27] zlibbioc_1.28.0 pkgbuild_1.0.2
## [29] Rhdf5lib_1.4.1 HDF5Array_1.10.0
## [31] DBI_1.0.0 rngtools_1.3.1
## [33] bibtex_0.4.2 xtable_1.8-3
## [35] progress_1.2.0 mclust_5.4.2
## [37] preprocessCore_1.44.0 httr_1.3.1
## [39] RColorBrewer_1.1-2 pkgconfig_2.0.2
## [41] reshape_0.8.8 XML_3.98-1.16
## [43] tidyselect_0.2.5 rlang_0.3.1
## [45] munsell_0.5.0 tools_3.5.3
## [47] cli_1.0.1 RSQLite_2.1.1
## [49] evaluate_0.12 stringr_1.3.1
## [51] processx_3.2.0 bit64_0.9-7
## [53] fs_1.2.6 beanplot_1.2
## [55] caTools_1.17.1.1 purrr_0.2.5
## [57] bindrcpp_0.2.2 nlme_3.1-137
## [59] doRNG_1.7.1 nor1mix_1.2-3
## [61] xml2_1.2.0 biomaRt_2.38.0
## [63] compiler_3.5.3 tibble_1.4.2
## [65] stringi_1.2.4 ps_1.2.1
## [67] desc_1.2.0 lattice_0.20-38
## [69] Matrix_1.2-16 multtest_2.38.0
## [71] pillar_1.3.0 data.table_1.11.8
## [73] bitops_1.0-6 rtracklayer_1.42.1
## [75] R6_2.3.0 KernSmooth_2.23-15
## [77] sessioninfo_1.1.1 codetools_0.2-15
## [79] assertthat_0.2.0 pkgload_1.0.2
## [81] rhdf5_2.26.0 openssl_1.1
## [83] pkgmaker_0.27 rprojroot_1.3-2
## [85] withr_2.1.2 GenomicAlignments_1.18.0
## [87] Rsamtools_1.34.0 GenomeInfoDbData_1.2.0
## [89] hms_0.4.2 quadprog_1.5-5
## [91] tidyr_0.8.2 base64_2.0
## [93] DelayedMatrixStats_1.4.0 base64enc_0.1-3
References
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.