,
David Corujo [aut],
Bernat Gel [aut]| Title: | RegioneReloaded: Multiple Association for Genomic Region Sets |
|---|---|
| Description: | RegioneReloaded is a package that allows simultaneous analysis of associations between genomic region sets, enabling clustering of data and the creation of ready-to-publish graphs. It takes over and expands on all the features of its predecessor regioneR. It also incorporates a strategy to improve p-value calculations and normalize z-scores coming from multiple analysis to allow for their direct comparison. RegioneReloaded builds upon regioneR by adding new plotting functions for obtaining publication-ready graphs. |
| Authors: | Roberto Malinverni [aut, cre] ,
David Corujo [aut],
Bernat Gel [aut] |
| Maintainer: | Roberto Malinverni <[email protected]> |
| License: | Artistic-2.0 |
| Version: | 1.9.0 |
| Built: | 2024-10-31 04:27:02 UTC |
| Source: | https://github.com/bioc/regioneReloaded |
The Alien Genome is an artificial genomic coordinates system for the purposes of testing and demonstrating the functions of regioneReload with a low computing time.
data(cw_Alien)data(cw_Alien)
An objects of class GRanges.
The Alien Genome consists of four chromosomes and is generated by the following code:
AlienGenome <-
toGRanges(data.frame(
chr = c("AlChr1", "AlChr2", "AlChr3", "AlChr4"),
start = c(rep(1, 4)),
end = c(2e6, 1e6, 5e5, 1e5)
))
List of region sets (as GRanges) on the AlienGenome.
data(cw_Alien)data(cw_Alien)
A list of GRanges objects.
This region sets are generated for the purpose of demonstrating the functions
of RegioneReloaded with a low computing time and "predictable" associations. The
regions are generated with by combining createRandomRegions() and
similarRegionSet() so that there is a known overlap between certain region
sets. To see a full description of this sample data and the code used to
generate it, see the RegioneReloaded vignette.
List of region sets (as GRanges) on the AlienGenome.
data(cw_Alien)data(cw_Alien)
A list of GRanges objects.
This region sets are generated for the purpose of demonstrating the functions
of RegioneReloaded with a low computing time and "predictable" associations. The
regions are generated with by combining createRandomRegions() and
similarRegionSet() so that there is a known overlap between certain region
sets. To see a full description of this sample data and the code used to
generate it, see the RegioneReloaded vignette.
Evaluate and choose the best method for clustering a matrix using the hclust() function.
chooseHclustMet(GM, scale = FALSE, vecMet = NULL, distHC = "euclidean")chooseHclustMet(GM, scale = FALSE, vecMet = NULL, distHC = "euclidean")
GM |
matrix, numerical matrix. |
scale |
logical, if TRUE, the clustering will be performed using the scaled matrix. (default = FALSE) |
vecMet |
character, vector of methods that will be tested in the function. If NULL, the following methods will be tested: "complete", "average", "single", "ward.D2", "median", "centroid" and "mcquitty. (default = NULL) |
distHC |
character, the distance measure to be used from those available in |
An object of class hclust
M1 <- matrix(1:18, nrow = 6, ncol = 3) set.seed(42) M2 <- matrix(sample(100, 18), nrow = 6, ncol = 3) GM <- cbind(M1, M2) chooseHclustMet(GM)M1 <- matrix(1:18, nrow = 6, ncol = 3) set.seed(42) M2 <- matrix(sample(100, 18), nrow = 6, ncol = 3) GM <- cbind(M1, M2) chooseHclustMet(GM)
Create the universe parameter for regioneR::resampleRegions() using all unique regions present in Alist.
createUniverse(Alist, joinR = TRUE)createUniverse(Alist, joinR = TRUE)
Alist |
list of regions set in a format accepted for regioneR |
joinR |
logical, if TRUE all the regions will be joined using the function |
A list of GRanges objects
data("cw_Alien") universe <- createUniverse(AlienRSList_narrow)data("cw_Alien") universe <- createUniverse(AlienRSList_narrow)
Perform multiple permutation tests between each element in two lists of region sets.
crosswisePermTest(Alist, Blist = NULL, sampling = FALSE, fraction = 0.15, min_sampling = 5000, ranFUN = "randomizeRegions", evFUN = "numOverlaps", ntimes = 100, universe = NULL, adj_pv_method = "BH", genome = "hg19", ...)crosswisePermTest(Alist, Blist = NULL, sampling = FALSE, fraction = 0.15, min_sampling = 5000, ranFUN = "randomizeRegions", evFUN = "numOverlaps", ntimes = 100, universe = NULL, adj_pv_method = "BH", genome = "hg19", ...)
Alist, Blist
|
GRangesList or list of region sets in any accepted formats by regioneR package (GRanges, data.frame etc.). |
sampling |
logical, if TRUE the function will use only a sample of each element of Alist to perform the test as specified in |
fraction |
logical, if |
min_sampling |
numeric, minimum number of regions accepted after sampling is performed with the specified |
ranFUN |
character, the randomization strategy used for the test, see regioneR. (default = "randomizeRegions") |
evFUN |
character, the evaluation strategy used for the test, see regioneR. (default = "numOverlaps) |
ntimes |
numeric, number of permutations used in the test. (default = 100) |
universe |
region set to use as universe, used only when |
adj_pv_method |
character, the method used for the calculation of the adjusted p-value, to choose between the options of |
genome |
character or GRanges, genome used to compute the randomization. (default = "hg19") |
... |
further arguments to be passed to other methods. |
This function performs multiple permutation tests for all pairwise combinations
of the elements in two lists of region sets. Essentially, it uses the regioneR::permTest()
function and its associated randomization and evaluation functions. It creates and returns a
genoMatriXeR object with the result of the permutation tests stored in the multiOverlaps slot.
In addition, all the parameters used for the test are stored in the parameters slot.
A object of class genoMatriXeR containing three slots
@parameters
@multioverlaps
@matrix
genoMatriXeR, regioneR, regioneR::permTest(), regioneR::overlapPermTest()
fakeGenome <- regioneR::toGRanges("chrF", 1, 1000) regA <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2, genome = fakeGenome) regB <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2, genome = fakeGenome) regAs <- similarRegionSet(GR = regA, genome = fakeGenome, name = "A", vectorPerc = seq(0.1, 0.3, by = 0.1)) regBs <- similarRegionSet(GR = regB, genome = fakeGenome, name = "B", vectorPerc = seq(0.1, 0.3, by = 0.1)) ABList <- c(regAs, regBs) cw_ptAB <- crosswisePermTest(ABList, genome = fakeGenome, ntimes = 10) print(cw_ptAB)fakeGenome <- regioneR::toGRanges("chrF", 1, 1000) regA <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2, genome = fakeGenome) regB <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2, genome = fakeGenome) regAs <- similarRegionSet(GR = regA, genome = fakeGenome, name = "A", vectorPerc = seq(0.1, 0.3, by = 0.1)) regBs <- similarRegionSet(GR = regB, genome = fakeGenome, name = "B", vectorPerc = seq(0.1, 0.3, by = 0.1)) ABList <- c(regAs, regBs) cw_ptAB <- crosswisePermTest(ABList, genome = fakeGenome, ntimes = 10) print(cw_ptAB)
Alien Genome crosswise matrix using regioneR::randomizeRegions , regioneR::circularRandomizeRegions, regioneR::resampleRegions, regioneR::resampleGenome functions as permutation strategies.
data(cw_Alien)data(cw_Alien)
An objects of class genoMatriXeR; see makeCrosswiseMatrix().
Alien Genome crosswise matrix using regioneR::randomizeRegions()
function a permutation strategy. Alist = AlienRSList_narrow, Blist = AlienRSList_narrow
data(cw_Alien)data(cw_Alien)
An objects of class genoMatriXeR; see makeCrosswiseMatrix().
Alien Genome crosswise matrix using regioneR::resampleGenome() function as permutations
trategy. Alist = AlienRSList_narrow, Blist = AlienRSList_narrow
data(cw_Alien)data(cw_Alien)
An objects of class genoMatriXeR; see makeCrosswiseMatrix().
Alien Genome crosswise matrix using regioneR::resampleGenome() function as permutations
trategy. Alist = AlienRSList_narrow, Blist = AlienRSList_broad
data(cw_Alien)data(cw_Alien)
An objects of class genoMatriXeR; see makeCrosswiseMatrix().
Alien Genome crosswise matrix using regioneR::resampleRegions() function a permutation
strategy.Alist = AlienRSList_narrow, Blist = AlienRSList_narrow
data(cw_Alien)data(cw_Alien)
An objects of class genoMatriXeR; see makeCrosswiseMatrix().
An S4 class for "genoMatriXeR" object.
parametersList of parameters used to create the object.
multiOverlapsResults of multiple pairwise permutation tests generated with crosswisePermTest().
matrixList of numerical matrices containing z-score, pvalues and correlation values generated with makeCrosswiseMatrix()
data("cw_Alien") AlienRSList_narrow_small <- AlienRSList_narrow[c("regA","regB","regC")] cw_test <- crosswisePermTest(Alist = AlienRSList_narrow_small,Blist = AlienRSList_narrow_small, sampling = FALSE, genome = AlienGenome, per.chromosome = TRUE, ranFUN = "resampleGenome", evFUN = "numOverlaps", ntimes = 10, mc.cores = 2) class(cw_test)data("cw_Alien") AlienRSList_narrow_small <- AlienRSList_narrow[c("regA","regB","regC")] cw_test <- crosswisePermTest(Alist = AlienRSList_narrow_small,Blist = AlienRSList_narrow_small, sampling = FALSE, genome = AlienGenome, per.chromosome = TRUE, ranFUN = "resampleGenome", evFUN = "numOverlaps", ntimes = 10, mc.cores = 2) class(cw_test)
get Object of class hclust from genoMatriXeR or multiLocalZScore
getHClust( rR, hctype = "rows")getHClust( rR, hctype = "rows")
rR |
A genoMatriXeR or multiLocalZScore object. |
hctype |
character. Can be "rows" or "cols". (default= "cols") |
an object of class hclust
genoMatriXeR, multiLocalZScore, hclust
data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) hc <- getHClust(cw_Alien_ReG) plot(hc)data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) hc <- getHClust(cw_Alien_ReG) plot(hc)
Returns the matrix from an genoMatriXeR or multiLocalZScore object.
getMatrix(rR)getMatrix(rR)
rR |
genoMatriXeR or multiLocalZScore object |
a numerical matrix from a
genoMatriXeR, multiLocalZScore, makeCrosswiseMatrix, makeLZMatrix
data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) mtx <- getMatrix(cw_Alien_ReG) mtx data("cw_Alien") cw_Alien_RaR <- makeCrosswiseMatrix(cw_Alien_RaR) GM <- getMatrix(cw_Alien_RaR) GMdata("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) mtx <- getMatrix(cw_Alien_ReG) mtx data("cw_Alien") cw_Alien_RaR <- makeCrosswiseMatrix(cw_Alien_RaR) GM <- getMatrix(cw_Alien_RaR) GM
Get multiEvaluation slot from genoMatriXeR or multiLocalZScore class.
getMultiEvaluation( rR, namesRS = NULL)getMultiEvaluation( rR, namesRS = NULL)
rR |
A genoMatriXeR or multiLocalZScore object. |
namesRS |
a vector of names. (default = NA) |
If rR is a genoMatriXeR object, a list of data frames resuming the associations results.
If rR is a multiLocalZScore object, a list of two elements: "resumeTable" that is a data frame
summarizing the associations and "shifts", a list of shifts computed from multiLocalZscore() function for the elements
indicated in the nameRS vector.
genoMatriXeR, multiLocalZScore
data("cw_Alien") mevs <- getMultiEvaluation(cw_Alien_ReG, names = "regA") mevsdata("cw_Alien") mevs <- getMultiEvaluation(cw_Alien_ReG, names = "regA") mevs
Get parameters from a genoMatriXeR or multiLocalZScore class object.
getParameters(rR, show_err = FALSE)getParameters(rR, show_err = FALSE)
rR |
A genoMatriXeR or multiLocalZScore class object. |
show_err |
logical, if TRUE the function returns a list with two dataframes: one containing the parameter values and one with any error messages that have been generated during the permutation test iterations when running crosswisePermTest. |
A dataframe with parameters and values, or a list with two dataframes with parameters and errors information.
genoMatriXeR, multiLocalZScore
data("cw_Alien") prm <- getParameters(cw_Alien_ReG) prmdata("cw_Alien") prm <- getParameters(cw_Alien_ReG) prm
Populate the matrix slot in a genoMatriXeR object.
makeCrosswiseMatrix(mPT, clusterize = TRUE, hc.method = NULL, dist.method = "euclidean", transform = FALSE, scale = FALSE, zs.type = 'norm_zscore', symm_matrix = TRUE, selectRow = NULL, selectCol = NULL, pvcut = 1, subEX = 0, GM_diag = TRUE, ...)makeCrosswiseMatrix(mPT, clusterize = TRUE, hc.method = NULL, dist.method = "euclidean", transform = FALSE, scale = FALSE, zs.type = 'norm_zscore', symm_matrix = TRUE, selectRow = NULL, selectCol = NULL, pvcut = 1, subEX = 0, GM_diag = TRUE, ...)
mPT |
an object of class genoMatriXeR. |
clusterize |
logical, if TRUE the matrix will be clustered using the method specified by |
hc.method |
character, select the |
dist.method |
character, the distance measure to be used from those available in |
transform |
logical, if TRUE the matrix will be transformed using the function |
scale |
logical, if TRUE the matrix will be scaled. (default = FALSE) |
zs.type |
character, z-score type to use to generate the matrix, either raw z-score ("zscore") or normalized z-score ("norm_zscore"). (default = "norm_zscore") |
symm_matrix |
logical, if TRUE the matrix will be treated as symmetrical (same clustering for rows and columns). (default = TRUE) |
selectRow, selectCol
|
vector, the matrix will be reduced selecting the rows and/or columns in this vector. (default = NULL) |
pvcut |
numeric, the z-score value is substituted by |
subEX |
numeric, value used to substitute the z-score values when the associated pvalue is higher than |
GM_diag |
logic, if FALSE the values of the diagonal will be set to 0. (default = TRUE) |
... |
further arguments to be passed to other methods. |
This function will create a series of matrices of z-scores, adj.pvalues and
pearson correlation values from all the pairwise permutation tests stored in
the multiOverlaps slot of a genoMatriXeR as
calculated with multiPermTest(). These matrices will then be stored in the
matrix slot of the genoMatriXeR object. In addition,
clustering will be performed on the association matrices using hclust.
An object of class genoMatriXeR containing three slots, with a populated matrix slot.
@parameters
@multioverlaps
@matrix
crosswisePermTest(), chooseHclustMet(), plotCrosswiseMatrix()
data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) summary(cw_Alien_ReG)data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) summary(cw_Alien_ReG)
Create a local z-score matrix from a multiLocalZScore
object and save it in its matrix slot.
makeLZMatrix(mlZA, normalize = TRUE, clusterize = TRUE, centralize = NA, hc.method = NULL, dist.method = "euclidean", scale = FALSE, ...)makeLZMatrix(mlZA, normalize = TRUE, clusterize = TRUE, centralize = NA, hc.method = NULL, dist.method = "euclidean", scale = FALSE, ...)
mlZA |
an object of class multiLocalZScore or a numerical matrix. |
normalize |
logical, if TRUE the z-score values in the matrix will be normalized. (default = FALSE) |
clusterize |
logical, if TRUE the matrix will be clustered using the method specified by |
centralize |
numeric, only z-score values in a number of steps (defined by |
hc.method |
character, select the |
dist.method |
character, the distance measure to be used from those available in |
scale |
logical, if TRUE the matrix will be scaled. (default = FALSE) |
... |
further arguments to be passed to other methods. |
A object of class multiLocalZScore containing three slots, with a populated matrix slot.
@parameters
@multiLocalZscores
@matrix
data("cw_Alien")data("cw_Alien")
Alien Genome multiLocalZScore calculated for regA regionset from AlienRSList_narrow using regioneR::resampleGenome() function as permutation s
trategy.
data(cw_Alien)data(cw_Alien)
An objects of class multiLocalZScore; see makeLZMatrix().
Alien Genome multiLocalZScore calculated for regA regionset from AlienRSList_broad using regioneR::resampleGenome() function as permutation s
trategy.
data(cw_Alien)data(cw_Alien)
An object of class multiLocalZScore
Alien Genome multiLocalZScore calculated for regD regionset from AlienRSList_narrow using regioneR::resampleGenome() function as permutation s
trategy.
data(cw_Alien)data(cw_Alien)
An object of class multiLocalZScore
Perform multiple permutation tests between a region set and each element in a list of region sets using shifted positions to calculate a local z-score.
multiLocalZscore(A, Blist = NULL, sampling = FALSE, fraction = 0.15, min_sampling = 5000, ranFUN = "randomizeRegions", evFUN = "numOverlaps", ntimes = 100, adj_pv_method = "BH", genome = "hg19", universe = NULL, window = 1000, step = 100, ...)multiLocalZscore(A, Blist = NULL, sampling = FALSE, fraction = 0.15, min_sampling = 5000, ranFUN = "randomizeRegions", evFUN = "numOverlaps", ntimes = 100, adj_pv_method = "BH", genome = "hg19", universe = NULL, window = 1000, step = 100, ...)
A |
query region set for which to estimate local z-score values. |
Blist |
GRangesList or list of region sets in any accepted formats by regioneR package (GRanges, data.frame etc.). |
sampling |
logical, if TRUE the function will use only a sample of each element of Alist to perform the test as specified in |
fraction |
logical, if |
min_sampling |
numeric, minimum number of regions accepted after sampling is performed with the specified |
ranFUN |
character, the randomization strategy used for the test, see regioneR. (default = "randomizeRegions") |
evFUN |
character, the evaluation strategy used for the test, see regioneR. (default = "numOverlaps) |
ntimes |
numeric, number of permutations used in the test. (default = 100) |
adj_pv_method |
character, the method used for the calculation of the adjusted p-value, to choose between the options of |
genome |
character or GRanges, genome used to compute the randomization. (default = "hg19") |
universe |
region set to use as universe, used only when |
window |
numeric, window (number of base pairs) in which the local z-score will be calculated. (default = 1000) |
step |
numeric, step (number of base pairs) by which will be estimated the local Z-score. (default = 100) |
... |
further arguments to be passed to other methods. |
This function performs multiple permutation tests between a single region set
and each element in a list of region sets. For every pairwise combination, the
evaluation step is repeated each time shifting the position of all the regions in the query region set
by a fixed step inside a defined window (using regioneR::localZScore().
This produces a "local z-score" profile that can be indicative of the nature
of the association between region sets. For example, an association can occur
"centrally" if the z-score value drops sharply when sifting the region set.
On the other hand, two region sets may have a peak of local z-score away from
the central position if they happen to occur often at a regular distance,
showing a "lateral" association.
A object of class multiLocalZScore containing three slots
@parameters
@multiLocalZscores
@matrix
fakeGenome<- regioneR::toGRanges("chrF",1,1000) regA <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2,genome = fakeGenome) regB <- regioneR::createRandomRegions(nregions = 10,length.mean = 10, length.sd = 2,genome = fakeGenome) regAs <-similarRegionSet(GR = regA,genome = fakeGenome, name = "A", vectorPerc = seq(0.1,0.3,by =0.1)) regBs <-similarRegionSet(GR = regB,genome = fakeGenome, name = "B", vectorPerc = seq(0.1,0.3,by =0.1)) ABList <- c(regAs,regBs) mlz_ptAB <- multiLocalZscore(A = regA, Blist = ABList, genome = fakeGenome, ntimes = 10) summary(mlz_ptAB)fakeGenome<- regioneR::toGRanges("chrF",1,1000) regA <- regioneR::createRandomRegions(nregions = 10, length.mean = 10, length.sd = 2,genome = fakeGenome) regB <- regioneR::createRandomRegions(nregions = 10,length.mean = 10, length.sd = 2,genome = fakeGenome) regAs <-similarRegionSet(GR = regA,genome = fakeGenome, name = "A", vectorPerc = seq(0.1,0.3,by =0.1)) regBs <-similarRegionSet(GR = regB,genome = fakeGenome, name = "B", vectorPerc = seq(0.1,0.3,by =0.1)) ABList <- c(regAs,regBs) mlz_ptAB <- multiLocalZscore(A = regA, Blist = ABList, genome = fakeGenome, ntimes = 10) summary(mlz_ptAB)
An S4 class for "multiLocalZScore" object.
parametersList of parameters used to create the object
multiLocalZscoresResults of multiple pairwise permutation tests on shifted region sets generated with multiLocalZscore().
matrixList of numerical matrices containing local z-scores and correlation values generated with makeLZMatrix().
data("cw_Alien") AlienRSList_narrow_small <- AlienRSList_narrow[c("regA","regB","regC")] mlz_test <- multiLocalZscore(A = AlienRSList_narrow_small$regA, Blist = AlienRSList_narrow_small, sampling = FALSE, genome = AlienGenome, per.chromosome = TRUE, ranFUN = "resampleGenome", evFUN = "numOverlaps", ntimes = 10, mc.cores = 2) class(mlz_test)data("cw_Alien") AlienRSList_narrow_small <- AlienRSList_narrow[c("regA","regB","regC")] mlz_test <- multiLocalZscore(A = AlienRSList_narrow_small$regA, Blist = AlienRSList_narrow_small, sampling = FALSE, genome = AlienGenome, per.chromosome = TRUE, ranFUN = "resampleGenome", evFUN = "numOverlaps", ntimes = 10, mc.cores = 2) class(mlz_test)
Plot a visualization of a genoMatriXeR object (or matrix) using different dimensional reduction algorithms (PCA, tSNE and UMAP).
plotCrosswiseDimRed(mPT, type = "PCA", GM_clust = NA, clust_met = "hclust", nc = 5, listRS = NULL, main = "", labSize = 2, emphasize = FALSE, labAll = FALSE, labMaxOverlap = 100, ellipse = TRUE, colPal = NULL, perplexity = 10, theta = 0.1, return_table = FALSE, return_plot = TRUE, ...)plotCrosswiseDimRed(mPT, type = "PCA", GM_clust = NA, clust_met = "hclust", nc = 5, listRS = NULL, main = "", labSize = 2, emphasize = FALSE, labAll = FALSE, labMaxOverlap = 100, ellipse = TRUE, colPal = NULL, perplexity = 10, theta = 0.1, return_table = FALSE, return_plot = TRUE, ...)
mPT |
an object of class genoMatriXeR or a numerical matrix. |
type |
character, dimensional reduction algorithm to use ("PCA", "tSNE", "UMAP"). (default = "PCA") |
GM_clust |
numeric, vector of assigned clusters used to cluster the matrix. If NA, the matrix will be clustered using the method defined by |
clust_met |
character, unsupervised cluster strategy used (hclust, kmeans or pam). (default = "hclust") |
nc |
numeric, number of clusters to define if using the default "kmeans" method. (default = 5) |
listRS |
list, a list of names of region sets of interest to be highlighted in the graph. (default = NULL) |
main |
character, title for the plot. (default = "") |
labSize |
numeric, size for point labels in the plot. If 0, no labels will be plotted. (default = 2) |
emphasize |
logical, if TRUE, only the cluster in which the elements of |
labAll |
logical, if TRUE all data points are labelled, even if not in |
labMaxOverlap |
numeric, max.overlaps for |
ellipse |
logical, if TRUE ellipses will be drawn around the clusters. (default = FALSE) |
colPal |
character, colors to use as palette for the plot. If NULL, default colors will be used. (default = NULL) |
perplexity, theta
|
numeric, if |
return_table |
logical, if TRUE a table with the cluster assigned to each region is returned. (default = FALSE) |
return_plot |
logical, if TRUE a plot is returned. (default = TRUE) |
... |
further arguments to be passed on to other methods |
This function generates a plot with a two-dimensional representation of the association data stored in a genoMatriXeR object by using either PCA, tSNE or UMAP transformations of the data. This function incorporates a clustering step and allows to highlight specific region sets of interest and the clusters they belong to. In addition to generating a plot, a table with the cluster assignments can be retrieved.
A ggplot object or a table with cluster assignments is returned.
data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) plotCrosswiseDimRed(cw_Alien_ReG, type = "PCA") CDR_clust <- plotCrosswiseDimRed(cw_Alien_ReG, type = "UMAP", return_table = TRUE) print(CDR_clust)data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix(cw_Alien_ReG) plotCrosswiseDimRed(cw_Alien_ReG, type = "PCA") CDR_clust <- plotCrosswiseDimRed(cw_Alien_ReG, type = "UMAP", return_table = TRUE) print(CDR_clust)
Plot matrix of associations/correlations stored in a genoMatriXeR object.
plotCrosswiseMatrix(mPT, lineColor = NA, interpolate = FALSE, colMatrix = "default", matrix_type = "association", cor = "row", maxVal = NA, main = "", ord_mat = NULL)plotCrosswiseMatrix(mPT, lineColor = NA, interpolate = FALSE, colMatrix = "default", matrix_type = "association", cor = "row", maxVal = NA, main = "", ord_mat = NULL)
mPT |
an object of class genoMatriXeR or a numerical matrix. |
lineColor |
logical, color for the line grid delineating the tiles of the matrix plot. If NA, no line will be drawn. (default = NA) |
interpolate |
logical, if TRUE the image will be interpolated using the function |
colMatrix |
character or vector of colors, if "default" will be used a default selection see.. |
matrix_type |
character, type of matrix to be plotted, either "association" or "correlation". (default = "association") |
cor |
character, if |
maxVal |
numeric, maximum absolute value displayed by the plot. If "max", the maximum values in the matrix are used. If NA, the 0.95 quantile of all absolute values is used. (default = NA) |
main |
character, title of the plot. (default = "") |
ord_mat |
numeric, list with two numeric vectors that represent the ordering of rows and column of the matrix to be used in the plot. If NULL, the order of the matrix is preserved as is. (default = NULL) |
This functions creates a graphical representation of the matrix of associations stored in a genoMatriXeR object. The values plotted and clustering options can be controlled when creating the matrix with the function makeCrosswiseMatrix.
Returns a ggplot object.
crosswisePermTest makeCrosswiseMatrix
data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix( cw_Alien_ReG) plotCrosswiseMatrix(cw_Alien_ReG, matrix_type = "association") plotCrosswiseMatrix(cw_Alien_ReG, matrix_type = "correlation")data("cw_Alien") cw_Alien_ReG <- makeCrosswiseMatrix( cw_Alien_ReG) plotCrosswiseMatrix(cw_Alien_ReG, matrix_type = "association") plotCrosswiseMatrix(cw_Alien_ReG, matrix_type = "correlation")
Plot Local Z-Score Matrix of associations/correlations stored in a multiLocalZScore object.
plotLocalZScoreMatrix (mLZ, lineColor = NA, colMatrix = "default", matrix_type = "association", maxVal = "max", main = "", labSize = 6, revert = FALSE, highlight = NULL, highlight_size = 2.5, highlight_max = FALSE, smoothing = FALSE, ...)plotLocalZScoreMatrix (mLZ, lineColor = NA, colMatrix = "default", matrix_type = "association", maxVal = "max", main = "", labSize = 6, revert = FALSE, highlight = NULL, highlight_size = 2.5, highlight_max = FALSE, smoothing = FALSE, ...)
mLZ |
an object of class multiLocalZScore or a matrix |
lineColor |
logical, color for the line grid delineating the tiles of the matrix plot. If NA, no line will be drawn. (default = NA) |
colMatrix |
character or vector of colors, if "default" will be used a default selection see.. |
matrix_type |
character, type of matrix to be plotted, either "association" or "correlation". (default = "association") |
maxVal |
numeric, maximum absolute value displayed by the plot. If "max", the maximum values in the matrix are used. If NA, the 0.95 quantile of all absolute values is used. (default = NA) |
main |
character, title of the plot. (default = "") |
labSize |
numeric, size for the plot labels. (default = 6) |
revert |
logical, if TRUE reverts the order of the plotted elements. (default = FALSE) |
highlight |
character, vector indicating the region set names to highlight by adding labels pointing to the 0 shift position (default = NULL) |
highlight_size |
numeric, size of the highlight labels. (default = 2.5) |
highlight_max |
logical, if TRUE the highlight labels are placed at the maximum local z-score value instead of the 0 shift position. (default = FALSE) |
smoothing |
logical, if TRUE the stats::smooth.spline function will be applied to the local z-score profile. (default = FALSE) |
... |
further arguments to be passed to other methods. |
Returns a ggplot object.
multiLocalZscore makeLZMatrix multiLocalZScore
data("cw_Alien")data("cw_Alien")
Plot the result of specific local Z-Score tests from a multiLocalZScore object in the form of line plot profiles.
plotSingleLZ(mLZ, RS, xlab = "", normZS = TRUE, ylim = NULL, main = NA, colPal = NULL, labValues = TRUE, labSize = 2.5, labMax = FALSE, smoothing = FALSE, ...)plotSingleLZ(mLZ, RS, xlab = "", normZS = TRUE, ylim = NULL, main = NA, colPal = NULL, labValues = TRUE, labSize = 2.5, labMax = FALSE, smoothing = FALSE, ...)
mLZ |
an object of class multiLocalZScore. |
RS |
character, vector of region set names for which to plot the local Z-score results. |
xlab |
character, label for the x axis. (default = NA) |
normZS |
logical, indicates whether the normalized Z-score values should be plotted. If FALSE, the raw Z-score is used. (default = TRUE) |
ylim |
numeric, vector with two elements: minimum and maximum Y values of the plot. If NULL, the plot limits are set by default so all data points can be plotted. (default = NULL) |
main |
character, title for the plot. If NA, the name of the query region set in the multiLocalZScore object will be used. (default = NA) |
colPal |
character, colors to use as palette for the plot. If NULL, default colors will be used. (default = NULL) |
labValues |
logical, if TRUE each local Z-score profile is labelled at position 0 with the name of the region set and its Z-score value at the central position. (default = TRUE) |
labSize |
numerical, size of the labels from labValues in the plot. (default = 2.5) |
labMax |
logical, if TRUE the labels are placed at the maximum value of each local Z-score profile instead of the center. (default = FALSE) |
smoothing |
logical, if TRUE the |
... |
further arguments to be passed to other methods. |
This function generates a line plot with the local Z-score profiles of selected region sets from a multiLocalZScore object. This type of plot complements the local Z-score matrix (generated by plotLocalZScoreMatrix, since it allows to visualize in detail the local Z-score profile of just the region sets of interest.
This plot is well suited for a single or a few region sets, but will get busy if attempting to plot many different profiles. For the latter, the full matrix generated by plotLocalZScoreMatrix is usually a better visualization option.
Returns a ggplot object.
multiLocalZscore(), makeLZMatrix()
data("cw_Alien") plotSingleLZ(mLZ_regA_ReG, RS = c("regD", "regD_02", "regA", "regAB_04"), labMax = TRUE, smoothing = TRUE)data("cw_Alien") plotSingleLZ(mLZ_regA_ReG, RS = c("regD", "regD_02", "regA", "regAB_04"), labMax = TRUE, smoothing = TRUE)
Plot the result of a single pairwise permutation test from a genoMatriXeR object.
plotSinglePT(mPT, RS1, RS2, xlab = NA, main = NA)plotSinglePT(mPT, RS1, RS2, xlab = NA, main = NA)
mPT |
an object of class genoMatriXeR. |
RS1, RS2
|
character, names of region sets in a genoMatriXeR object for which to represent the pairwise permutation test results. |
xlab |
character, label for x axis. (default = NA) |
main |
title for the plot, if NA the name of the genoMatriXeR object is used (default = NA) |
This function generates a plot representing the result of a single permutation test stored in a genoMatriXeR object. This includes a plot of the density distribution of the randomized evaluations and a vertical line showing the observed evaluation in the original region set. The values of the mean randomized evaluations and the value of the observed evaluation are shown, in addition to the calculated Z-score, normalized Z-score and adjusted p-value.
Returns a ggplot object.
crosswisePermTest makeCrosswiseMatrix
data("cw_Alien") plotSinglePT(cw_Alien_ReG, RS1 = "regA", RS2 = "regA_05") plotSinglePT(cw_Alien_ReG, RS1 = "regA", RS2 = "regC")data("cw_Alien") plotSinglePT(cw_Alien_ReG, RS1 = "regA", RS2 = "regA_05") plotSinglePT(cw_Alien_ReG, RS1 = "regA", RS2 = "regC")
Create a random region set similar to a reference region set.
randomizeRegionsPerc(GR, genome = "hg19", frac = 0.2, ...)randomizeRegionsPerc(GR, genome = "hg19", frac = 0.2, ...)
GR |
a GRanges object with the input region set. |
genome |
genome of reference to generate the similar region sets. (default = "hg19) |
frac |
fraction of the original region set to randomize. (default = 0.2) |
... |
further arguments to be passed to other methods. |
This function takes an input region set and generates a region set where a fraction of the regions has been randomized.
a GRanges object
data("cw_Alien") nreg <- 100 regA <- createRandomRegions( nregions = nreg, length.mean = 100, length.sd = 10, non.overlapping = TRUE, genome = AlienGenome ) regA_02 <- randomizeRegionsPerc(GR = regA, genome = AlienGenome, frac = 0.2)data("cw_Alien") nreg <- 100 regA <- createRandomRegions( nregions = nreg, length.mean = 100, length.sd = 10, non.overlapping = TRUE, genome = AlienGenome ) regA_02 <- randomizeRegionsPerc(GR = regA, genome = AlienGenome, frac = 0.2)
Create a list of similar region sets to a reference region set.
similarRegionSet(GR, name, genome, vectorPerc)similarRegionSet(GR, name, genome, vectorPerc)
GR |
a GRanges object with the input region set. |
name |
character, name for the output region sets. The names will be generated by adding an underscore and the fraction of similarity after the name of each region set generated. (default = "A") |
genome |
genome of reference to generate the similar region sets. (default = "hg19) |
vectorPerc |
numeric, vector of desired randomized fractions. (default = seq(.1,.9,.1)) |
This function takes a region set as an input and a vector of desired randomized
fractions. For each fraction value, a new region set will be generated
where that fraction of the original regions in the input region set has been randomized.
In effect, this creates region sets that are "similar" to a controlled degree
to the original region set. This tool can be useful for validation purposes
and its use in the demonstration of the usage of this package can be
seen in the RegioneReloaded vignette.
A list of GRanges objects.
data("cw_Alien") A<-createRandomRegions(nregions = 20, length.mean = 1000, length.sd = 100, genome = AlienGenome) similAList <- similarRegionSet(GR = A, genome = AlienGenome, vectorPerc = seq(0.1,0.9,0.2), name = "test") summary (similAList) data("cw_Alien") regA <- createRandomRegions( nregions = 100, length.mean = 10, length.sd = 5, genome = AlienGenome ) listRegA <- similarRegionSet(GR = regA, genome = AlienGenome) summary(listRegA)data("cw_Alien") A<-createRandomRegions(nregions = 20, length.mean = 1000, length.sd = 100, genome = AlienGenome) similAList <- similarRegionSet(GR = A, genome = AlienGenome, vectorPerc = seq(0.1,0.9,0.2), name = "test") summary (similAList) data("cw_Alien") regA <- createRandomRegions( nregions = 100, length.mean = 10, length.sd = 5, genome = AlienGenome ) listRegA <- similarRegionSet(GR = regA, genome = AlienGenome) summary(listRegA)