start_decomp_pipeline: start_decomp_pipeline
In lutsik/DecompPipeline: Preparation pipeline for MeDeCom

Description Usage Arguments Value Author(s)

Main workhorse of the DecompPipeline R-package. Performs preprocessing (prepare_data or prepare_data_BS), CpG subset selection (prepare_CG_subsets) and deconvolution (start_medecom_analysis, start.refreeewas.analysis, start.edec.analysis)

start_decomp_pipeline(rnb.set, Ks, lambda.grid, work.dir = getwd(),
  factorviz.outputs = F, analysis.name = "Analysis",
  sample.selection.col = NA, sample.selection.grep = NA,
  pheno.cols = NA, id.column = rnb.getOption("identifiers.column"),
  normalization = "none", ref.ct.column = NA, ref.rnb.set = NULL,
  ref.rnb.ct.column = NA, prepare.true.proportions = F,
  true.A.token = NA, houseman.A.token = NA,
  estimate.houseman.prop = F,
  filter.beads = !is.null(rnb.set@covg.sites), min.n.beads = 3,
  filter.intensity = inherits(rnb.set, "RnBeadRawSet"),
  min.int.quant = 0.01, max.int.quant = 0.99, filter.na = TRUE,
  filter.context = TRUE, filter.cross.reactive = TRUE,
  execute.lump = FALSE, remove.ICA = FALSE, conf.fact.ICA = FALSE,
  ica.setting = NULL, filter.snp = TRUE, filter.somatic = TRUE,
  snp.list = NULL, filter.coverage = hasCovg(rnb.set),
  min.coverage = 5, min.covg.quant = 0.05, max.covg.quant = 0.95,
  marker.selection = "var", n.markers = 5000,
  remove.correlated = FALSE, cor.threshold = "quantile",
  write.files = FALSE, n.prin.comp = 10, range.diff = 0.05,
  custom.marker.file = "", store.heatmaps = F,
  heatmap.sample.col = NULL, sample.subset = NULL, k.fixed = NULL,
  K.prior = NULL, opt.method = "MeDeCom.cppTAfact", startT = NULL,
  startA = NULL, folds = 10, cores = 1, itermax = 1000,
  ninit = 100, cluster.submit = FALSE, cluster.Rdir = NA,
  cluster.hostlist = "*", cluster.memlimit = "5G", cleanup = FALSE)

`rnb.set`	An object of type `RnBSet-class` for which analysis is to be performed.
`Ks`	Vector of integers used as components in MeDeCom.
`lambda.grid`	Vector of doubles representing the regularization parameter in MeDeCom.
`work.dir`	A path to a existing directory, in which the results are to be stored
`factorviz.outputs`	Flag indicating, if outputs should be stored to be compatible with FactorViz for data exploration
`analysis.name`	A string representing the dataset for which analysis is to be performed. Only used to create a folder with a descriptive name of the analysis.
`sample.selection.col`	A column name in the phenotypic table of `RNB_SET` used to selected a subset of samples for analysis that contain the string given in `SAMPLE_SELECTION_GREP`.
`sample.selection.grep`	A string used for selecting samples in the column `SAMPLE_SELECTION_COL`.
`pheno.cols`	Vector of column names in the phenotypic table of `RNB_SET` that is kept and exported for further exploration.
`id.column`	Sample-specific ID column name in `RNB_SET`
`normalization`	Normalization method to be performed before employing MeDeCom. Can be one of `"none","dasen","illumina","noob"` (BeadChip only).
`ref.ct.column`	Column name in `RNB_SET` used to extract methylation information on the reference cell types.
`ref.rnb.set`	An object of type `RnBSet-class` containing methylation information on reference cell types (BeadChip only).
`ref.rnb.ct.column`	Column name in `REF_RNB_SET` used to extract methylation information on the reference cell types (BeadChip only).
`prepare.true.proportions`	Flag indicating if true proportions are either available in `RNB_SET` or to be estimated with Houseman's reference-based deconvolution approach (BeadChip only).
`true.A.token`	String present in the column names of `RNB_SET` used for selecting the true proportions of the corresponding cell types.
`houseman.A.token`	Similar to `TRUE_A_TOKEN`, but not containing the true proportions, rather the estimated proportions by Houseman's method (BeadChip only).
`estimate.houseman.prop`	If neither `TRUE_A_TOKEN` nor `HOUSEMAN_A_TOKEN` are given, the proportions of the reference cell type are estimated with Houseman's approach (BeadChip only).
`filter.beads`	Flag indicating, if site-filtering based on the number of beads available is to be conducted (BeadChip only).
`min.n.beads`	Minimum number of beads required in each sample for the site to be considered for adding to MeDeCom (BeadChip only).
`filter.intensity`	Flag indicating if sites should be removed according to the signal intensities (the lowest and highest quantiles given by `MIN_INT_QUANT` and `MAX_INT_QUANT`) (BeadChip only).
`min.int.quant`	Lower quantile of intensities which is to be removed (BeadChip only).
`max.int.quant`	Upper quantile of intensities which is to be removed (BeadChip only).
`filter.na`	Flag indicating if sites with any missing values are to be removed or not.
`filter.context`	Flag indicating if only CG probes are to be kept (BeadChip only).
`filter.cross.reactive`	Flag indicating if sites showing cross reactivity on the array are to be removed.
`execute.lump`	Flag indicating if the LUMP algorithm is to be used for estimating the amount of immune cells in a particular sample.
`remove.ICA`	Flag indicating if independent component analysis is to be executed to remove potential confounding factor. If `TRUE`,conf.fact.ICA needs to be specified.
`conf.fact.ICA`	Column name in the sample annotation sheet representing a potential confounding factor.
`ica.setting`	Optional argument setting up ICA.
`filter.snp`	Flag indicating if annotated SNPs are to be removed from the list of sites according to RnBeads' SNP list. (@TODO: we could provide an addititional list of SNPs, similar to RnBeads blacklist for filtering)
`filter.somatic`	Flag indicating if only somatic probes are to be kept. CPG FILTERING (BS)
`snp.list`	Path to a file containing CpG IDs of known SNPs to be removed from the analysis, if `FILTER_SNP` is `TRUE`.
`filter.coverage`	Flag indicating, if site-filtering based on coverage is to be conducted (BS only).
`min.coverage`	Minimum number of reads required in each sample for the site to be considered for adding to MeDeCom (BS only).
`min.covg.quant`	Lower quantile of coverages. Values lower than this value will be ignored for analysis (BS only).
`max.covg.quant`	Upper quantile of coverages. Values higher than this value will be ignored for analysis (BS only). CG_SUBSET SELECTION
`marker.selection`	A vector of strings representing marker selection methods. Available method are "`all`" Using all sites available in the input. "`pheno`" Selected are the top `N_MARKERS` site that differ between the phenotypic groups defined in data preparation or by `rnb.sample.groups`. Those are selected by employing limma on the methylation matrix. "`houseman2012`" The 50k sites reported as cell-type specific in the Houseman's reference- based deconvolution. See Houseman et.al. 2012. "`houseman2014`" Selects the sites said to be linked to cell type composition by `RefFreeEWAS`, which is similar to surrogate variable analysis. See Houseman et.al. 2014. "`jaffe2014`" The sites stated as related to cell-type composition Jaffe et.al. 2014. "`rowFstat`" Markers are selected as those found to be associated to the reference cell types with F-statistics. If this option is selected, `REF_DATA_SET` and `REF_PHENO_COLUMN` need to be specified. "`random`" Sites are randomly selected. "`pca`" Sites are selected as those with most influence on the principal components. "`var`" Selects the most variable sites. "`hybrid`" Selects (N_MARKERS/2) most variable and (N_MARKERS/2) random sites. "`range`" Selects the sites with the largest difference between minimum and maximum across samples. "`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. If a specific data set is to be provided, it needs to be done through `REF_DATA_SET`. "`custom`" Specifying a custom file with indices.
`n.markers`	The number of sites to be selected. Defaults to 5000.
`remove.correlated`	Flag indicating if highly correlated features are to be removed.
`cor.threshold`	Numeric indicating a threshold above which sites are not to be considered in the feature selection. If `"quantile"`, sites correlated higher than the 95th quantile are removed.
`write.files`	Flag indicating if the selected sites are to be stored on disk.
`n.prin.comp`	Optional argument deteriming the number of prinicipal components used for selecting the most important sites.
`range.diff`	Optional argument specifying the difference between maximum and minimum required.
`custom.marker.file`	Optional argument containing a file that specifies the indices used for employing MeDeCom.
`store.heatmaps`	Flag indicating if a heatmap of the selected input sites is to be create from the input methylation matrix. The files are then stored in the 'heatmaps' folder in WD.
`heatmap.sample.col`	Column name in the phenotypic table of `rnb.set`, used for creating a color scheme in the heatmap.
`sample.subset`	Vector of indices of samples to be included in the analysis. If `NULL`, all samples are included.
`k.fixed`	Columns in the T matrix that should be fixed. If `NULL`, no columns are fixed.
`K.prior`	K determined from visual inspection. Only has an influence, if `MARKER_SELECTION="pcadapt"`.
`opt.method`	Optimization method to be used. Either MeDeCom.quadPen or MeDeCom.cppTAfact (default).
`startT`	Inital matrix for T.
`startA`	Initial matrix for A.
`folds`	Integer representing the number of folds used in the analysis.
`cores`	Integer representing the number of cores to be used in the analysis.
`itermax`	Maximum number of iterations
`ninit`	Number if initialtions.
`cluster.submit`	Flag indicating, if the jobs are to be submitted to a scientific compute cluster (only SGE supported).
`cluster.Rdir`	Path to an executable version of R.
`cluster.hostlist`	Regular expression, on which basis hosts are selected in the cluster environment.
`cluster.memlimit`	the `memlimit` resource value of the cluster submission.
`cleanup`	Flag indicating if temprary files are to be deleted.