dba.analyze: Perform differential binding affinity analysis
In DiffBind: Differential Binding Analysis of ChIP-Seq Peak Data

Description Usage Arguments Details Value Note Author(s) See Also Examples

Performs differential binding affinity analysis. Performs default generation of a consensus peakset, read counting, normalization, and setting up of contrasts if they have not been specified.

dba.analyze(DBA, method=DBA$config$AnalysisMethod, design,
            bBlacklist=DBA$config$doBlacklist,
            bGreylist=DBA$config$doGreylist,
            bRetrieveAnalysis=FALSE, bReduceObjects=TRUE, 
            bParallel=DBA$config$RunParallel)

`DBA`	Either a `DBA` object, or a sample sheet (either a `character` vector with the name of the sample sheet, or a `data.frame` containing the experimental metadata. If no blacklist or greylists are included, a call will be made to `dba.blacklist` using defaults. This can be skipped by setting the `bBlacklist` and/or `bGreylist` parameters. If no counts are included, a default consensus will be formed and read counts computed via a call to `dba.count` using defaults. If no normalization has been specified, the reads will be normalized via a call to `dba.normalize` using defaults. If no contrasts are specified (`DBA$contrast` is `NULL`), default contrasts will be added via a call to `dba.contrast` using defaults.
`method`	Underlying method, or vector of methods, by which to analyze differential binding affinity. Supported methods: `DBA_EDGER` use `edgeR` package for analysis `DBA_DESEQ2` use `DESeq2` package for analysis `DBA_ALL_METHODS` perform two analyses, using both `edgeR` and `DESeq2`
`design`	If present and a character string, will be used as the design formula for the analysis, replacing any previously established design if present. If `FALSE`, will complete analysis in pre-version 3 mode. See `link{dba.contrast}`.
`bBlacklist`	If `TRUE`, and no blacklist has been applied to the `DBA` object (or when starting from a samplesheet), the read bam files will be examined to determine the reference genome, and an appropriate blacklist applied, if available. See `link{dba.blacklist}`.
`bGreylist`	If `TRUE`, and no greylist has been applied to the `DBA` object (or when starting from a samplesheet), the control bam files, if present, will be examined to determine the reference genome, greylists will be computed for each, merged into a master greylist, and applied to the peaksets. See `link{dba.blacklist}`.
`bRetrieveAnalysis`	If changed from `FALSE`, the underlying DE analysis object is returned instead of running a new analysis. Possible values for `bRetrieveAnalysis`: `DBA_DESEQ2` Returns the `DESeq2` `DESeqDataSet`. `DBA_EDGER` Returns the `edgeR` `DGEList`. `TRUE` Returns the `DESeq2` `DESeqDataSet`, if present. If not, returns the `edgeR` `DGEList`, if present.. An analysis object will only be successfully returned if there is at least one contrast utilizing an explicit design (see `dba.contrast`), and an analysis has been carried out.
`bReduceObjects`	logical indicating whether strip the analysis objects of unnecessary fields to save memory. If it is desired to use the `DBA$contrasts[[n]]$edgeR` and/or `DBA$contrasts[[n]]$DESeq2` objects directly in the `edgeR` and/or `DESeq2` packages, this should be set to `FALSE`.
`bParallel`	logical indicating that the analyses is to be done in parallel using multicore (one process for each contrast for each method, plus an additional process per method).

In general, prior to calling dba.analyze, dba.count should have been run. If no contrasts have been established prior to invoking dba.analyze, then the default set of contrasts will be added using (dba.contrast).

If no normalization parameters have been supplied by calling dba.normalize, default normalization parameters will be used.

See the DBA User Guide for more details on how the edgeR and DESeq2 analyses are carried out.

DBA object with results of analysis added to DBA$contrasts.

Alternatively, an analysis object (either a DESeqDataSet or a DGEList) if bRetrieveAnalysis if not FALSE.

If there is a blocking factor for the contrast(s) specified using a previous call to dba.contrast with design=FALSE, a multi-factor analysis will automatically be carried out in addition to a single factor analysis.

Rory Stark

dba.blacklist, dba.count, dba.contrast, dba.normalize, dba.report

data(tamoxifen_counts)
dba.analyze(tamoxifen)

tamoxifen <- dba.analyze(tamoxifen, method=DBA_ALL_METHODS,
                         design="~Tissue + Condition")
dba.show(tamoxifen, bContrasts=TRUE)

dba.analyze(tamoxifen, bRetrieveAnalysis=TRUE)
edger.object <- dba.analyze(tamoxifen, bRetrieveAnalysis=DBA_EDGER)
class(edger.object)