buildPEPs: Build PEPs from GEPs and stores them in the repository.
In gep2pep: Creation and Analysis of Pathway Expression Profiles (PEPs)

Description Usage Arguments Details Value See Also Examples

Given a matrix of ranked lists of genes (GEPs) and a gep2pep repository, converts GEPs to PEPs and stores the latter in the repository.

buildPEPs(rp, geps, min_size = 3, max_size = 500, parallel = FALSE,
  collections = "all", replace_existing = FALSE, donotstore = FALSE,
  progress_bar = TRUE, rawmode_id = NULL,
  rawmode_outdir = file.path(rp$root(), "raw"))

`rp`	A repository created by `createRepository`.
`geps`	A matrix of ranks where each row corresponds to a gene and each column to a condition. Each column must include all ranks from 1 to the number of rows. Row and column names must be defined. Row names will be matched against gene identifiers in the pathways collections, and unrecognized gene names will not be used.
`min_size`	An integer representing the minimum number of genes that must be included in a set before the KS statistic is computed. Smaller gene sets will get ES=NA and p=NA. Default is 3. Ignored for SGE mode (see `addSingleGeneSets`).
`max_size`	An integer representing the maximum number of genes that must be included in a set before the KS statistic is computed. Larger gene sets will get ES=NA and p=NA. Default is 500.
`parallel`	If TRUE, gene sets will be processed in parallel. Requires a parallel backend.
`collections`	A subset of the collection names returned by `getCollections`. If set to "all" (default), all the collections in `rp` will be used.
`replace_existing`	What to do if PEPs, identified by column names of `geps` are already present in the repository. If set to TRUE, they will be replaced, otherwise they will be skipped and only PEPs of new conditions will be computed and added. Either ways, will throw a warning.
`donotstore`	Just compute and return the pathway-based profiles without storing them in the repository. The repository is still required to load pathway data, however it will not be modified.
`progress_bar`	If set to TRUE (default) will show a progress bar updated after coversion of each column of `geps`.
`rawmode_id`	An integer to be appended to files produced in raw mode (see details). If set to NULL (default), raw mode is turned off.
`rawmode_outdir`	A charater vector specifying the destination path for files produced in raw mode (by the fault it is ROOT/raw, where ROOT is the root of the repository). Ignored if `rawmode_id` is NULL.

By deault, output is written to the repository as new items named using the collection name. However, it is possible to avoid the repository and write the output to regular files turning 'raw mode' on through the rawmode_id and rawmode_outdir parameters. This is particuarly useful when dealing with very large corpora of GEPs, and conversions are split into independent jobs submitted to a scheduler. At the end, the data will need to be reconstructed and put into the repository using importFromRawMode in order to perform CondSEA or PathSEA analysis.

Nothing. The computed PEPs will be available in the repository.

buildPEPs

db <- loadSamplePWS()
repo_path <- file.path(tempdir(), "gep2pepTemp")

rp <- createRepository(repo_path, db)
## Repo root created.
## Repo created.
## [15:45:06] Storing pathway data for collection: c3_TFT
## [15:45:06] Storing pathway data for collection: c3_MIR
## [15:45:06] Storing pathway data for collection: c4_CGN

rp
##          ID   Dims     Size
## c3_TFT_sets   10   18.16 kB
## c3_MIR_sets   10   17.25 kB
## c4_CGN_sets   10     6.9 kB

## Loading sample gene expression profiles
geps <- loadSampleGEP()

geps[1:3,1:3]
##       (+)_chelidonine (+)_isoprenaline (+/_)_catechin
## AKT3               88              117            417
## MED6              357              410             34
## NR2E3             383              121            453

buildPEPs(rp, geps)

rp
##           ID  Dims     Size
##   c3_TFT_sets   10 18.16 kB
##   c3_MIR_sets   10 17.25 kB
##   c4_CGN_sets   10   6.9 kB
##        c3_TFT    2  1.07 kB
##        c3_MIR    2  1.07 kB
##        c4_CGN    2  1.04 kB

unlink(repo_path, TRUE)