Condition Set Enrichment Analysis (CondSEA) can be seen as a Gene-SEA performed over rows (as opposed to columns) of a matrix of GEPs. It tells how much a pathway is consistently dysregulated under a set of conditions (such as a set of drug treatments, disease states, cell types, etc.) when compared to a statistical background of other conditions.
1 2 3
A repository created with
A vector of names of conditions. Corresponding PEPs
must exist in all the pathway collections currently in
The background against which to compare
A subset of the collection names returned by
If TRUE (default) rank details will be reported for
each condition in
The function used to rank PEPs column-wise. By
If set to TRUE, the computed ranked matrix will be stored in the the repository (see details). FALSE by default.
If TRUE (default) the output gene sets will be sorted in order of increasing p-value.
For each pathway, all conditions are ranked by how much
they dysregulate it (from the most UP-regulating to the most
DOWN-regulating). Then, a Kolmogorov-Smirnov (KS) test is performed
to compare the ranks assigned to conditions in
the ranks assigned to conditions in
bgset. A positive
(negative) Enrichment Score (ES) of the KS test indicates whether
each pathway is UP- (DOWN-) regulated by
pgset as compared
bgset. A p-value is associated to the ES.
When PEPs are obtained from drug-induced gene expression profiles,
PathSEA is the Drug-Set Enrichment Analysis .
rankingFun must take in input PEPs like those loaded
from the repository and return a matrix of row-wise ranks. Each row
must contains ranks from 1 to the number of PEPs minus the number
of NAs in the row.
usecache=TRUE, the ranked matrix is permanently stored
in HDF5 format, and subsequent calls to
CondSEA will load
from the disk the necessary ranks (not the whole matrix). The
correct cached data is identified by the alphabetically sorted set
union(pgset, bgset), by the collection name, and by the
ranking function. Additional alls to CondSEA with variations of
these inputs will create additional cache. Cached data is hidden in
the repository by default and can be printed with
rp_peps$print(all=TRUE), and cleared with
A list of 2, by names "CondSEA" and "details". The
"CondSEA" entry is a 2-columns matrix including ESs and p-values
(see details) for each pathway database and condition. The
"details" entry reports the rank of each condition in
for each pathway.
 Napolitano F. et al, Drug-set enrichment analysis: a novel tool to investigate drug mode of action. Bioinformatics 32, 235-241 (2016).
getResults, getDetails, clearCache
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
db <- loadSamplePWS() repo_path <- file.path(tempdir(), "gep2pepTemp") rp <- createRepository(repo_path, db) geps <- loadSampleGEP() buildPEPs(rp, geps) pgset <- c("(+)_chelidonine", "(+/_)_catechin") psea <- CondSEA(rp, pgset) res <- getResults(psea, "c3_TFT") ## getting the names of the top pathways setId2setName(loadCollection(rp, "c3_TFT"), rownames(res)) unlink(repo_path, TRUE)
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.