| Title: | The removal of batch effects from datasets using a PCA and constrained optimisation based technique |
|---|---|
| Description: | Harman is a PCA and constrained optimisation based technique that maximises the removal of batch effects from datasets, with the constraint that the probability of overcorrection (i.e. removing genuine biological signal along with batch noise) is kept to a fraction which is set by the end-user. |
| Authors: | Yalchin Oytam [aut], Josh Bowden [aut], Jason Ross [aut, cre] |
| Maintainer: | Jason Ross <[email protected]> |
| License: | GPL-3 + file LICENCE |
| Version: | 1.33.0 |
| Built: | 2024-07-02 05:38:45 UTC |
| Source: | https://github.com/bioc/Harman |
Generates an arrow plot for an instance of
harmanresults. The tail of the arrow is the starting point
(original) in principle coordinates, while the arrow head is the new point
(corrected) in principle coordinates. It can be observed that on principle
components that have undergone correction
(harmanresults$stats$correction < 1.0), the samples within a batch
will be coordinately moved towards 0 on that principle component.
arrowPlot( harmanresults, pc_x = 1, pc_y = 2, colBy = "batch", palette = "rainbow", col, length = 0.1, legend = TRUE, ... )arrowPlot( harmanresults, pc_x = 1, pc_y = 2, colBy = "batch", palette = "rainbow", col, length = 0.1, legend = TRUE, ... )
harmanresults |
an instance of |
pc_x |
integer, principle component for the plot x dimension. |
pc_y |
integer, principle component for the plot y dimension. |
colBy |
string, colour the points by the experimental or batch
variable; legal values are |
palette |
string, the function to call to create a vector of
contiguous colours with the levels of factor in |
col |
colour vector for the points. This parameter overrides
|
length |
length of the |
legend |
logical, whether to display a legend on the plot |
... |
further arguments passed to or from other methods. |
Generates a Principle Component plot for an instance of
harmanresults. If a vector of colours is supplied via the col
argument, then a legend will not be drawn.
None
harmanresults plot.harmanresults
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) arrowPlot(olf.harman, pc_x=2, pc_y=3, length=0.2)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) arrowPlot(olf.harman, pc_x=2, pc_y=3, length=0.2)
This wrapper should probably not be addressed directly except
for debugging. Instead use harman. Input of PCA scores and the
experiment structure (treatments and batches) and returns a batch corrected
version of the PCA scores matrix.
.callHarman( pc_data_scores, group, limit, numrepeats, randseed, forceRand, printInfo ).callHarman( pc_data_scores, group, limit, numrepeats, randseed, forceRand, printInfo )
pc_data_scores |
2D NumericMatrix of PCA scores data (from the
|
group |
The structure of the experiment, consisting of batch numbers and treatment numbers forming 2 rows or columns (HarmanMain works out which). Each entry for a sample describes what batch it came from and what treatment it was given. Has to be integer formated data. |
limit |
A double precsion value indicating the limit of confidence in which to stop removing a batch effect |
numrepeats |
The number of repeats in which to run the simulated batch mean distribution estimator. Probably should be greater than 100,000. |
randseed |
Random seed to pass to the random number generator (0 for use default from system time) |
forceRand |
Force algorithm |
printInfo |
Print update information to screen |
SEXP R list
scores.corrected
correction
confidence
A data matrix with samples in columns must be transposed before PCA
analysis and these scores in turn are tweaked a little before handing over
to .callHarman. See the example below.
This function is part of a set of three functions to be run in
series. discoverClusteredMethylation takes a matrix of
methylation beta values (typically from the Illumina Infinium Methylation
Assay) and clusters the data across a range of ks specified by the user.
Then the data is reclustered again across the the best two candidate values for k (determined by the rate of change in Bayesian information criterion), and minimum cluster size and distance filters are employed. If both clusters meet these filters, then the higher value of k is returned. This function should be run on uncorrected data that ideally has slides removed which are prone to batch effect. This will bias towards finding clusters that are driven by biological factors such as X-chromosome inactivation and allele-specific methylation.
The output of this function is input for the
kClusterMethylation function which extracts cluster membership
and statistics on variance for a given matrix of beta values. It might be
useful to discover clusters on samples less prone to clustering due to batch
effect or cellular heterogeneity and then recluster all the data for set
values of k via the kClusterMethylation function.
Finally, a comparison of differences of uncorrected to
batch-corrected beta values can be made using clusterStats.
This function generates a data.frame containing log variance ratio and mean
beta differences to clusters after correction.
clusterStats(pre_betas, post_betas, kClusters)clusterStats(pre_betas, post_betas, kClusters)
pre_betas |
a matrix of methylation beta values prior to correction. |
post_betas |
a matrix of methylation beta values after correction. |
kClusters |
a kClusters S3 object |
Betas values should be of type double with samples in
columns and betas in rows. The betas need to be bounded between 0 and 1.
The matrix is typically exported from a GenomicRatioSet,
GenomicMethylSet or MethylSet
object via the getBeta S4 accessor method.
A data.frame containing clustering stats.
kClusterMethylation,
discoverClusteredMethylation
library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)
A helper function that can be called if harman
had to be aborted.
detachHarman()detachHarman()
None
This function is part of a set of three functions to be run in
series. discoverClusteredMethylation takes a matrix of
methylation beta values (typically from the Illumina Infinium Methylation
Assay) and clusters the data across a range of ks specified by the user.
Then the data is reclustered again across the the best two candidate values for k (determined by the rate of change in Bayesian information criterion), and minimum cluster size and distance filters are employed. If both clusters meet these filters, then the higher value of k is returned. This function should be run on uncorrected data that ideally has slides removed which are prone to batch effect. This will bias towards finding clusters that are driven by biological factors such as X-chromosome inactivation and allele-specific methylation.
The output of this function is input for the
kClusterMethylation function which extracts cluster membership
and statistics on variance for a given matrix of beta values. It might be
useful to discover clusters on samples less prone to clustering due to batch
effect or cellular heterogeneity and then recluster all the data for set
values of k via the kClusterMethylation function.
Finally, a comparison of differences of uncorrected to
batch-corrected beta values can be made using clusterStats.
This function generates a data.frame containing log variance ratio and mean
beta differences to clusters after correction.
discoverClusteredMethylation( betas, ks = 1:10, min_cluster_size = 5, min_cluster_dist = 0.1, max_clusters = 4, cores = 1, printInfo = FALSE )discoverClusteredMethylation( betas, ks = 1:10, min_cluster_size = 5, min_cluster_dist = 0.1, max_clusters = 4, cores = 1, printInfo = FALSE )
betas |
a matrix of methylation beta values with samples in columns and betas in rows. |
ks |
integer, the range of k's to consider for clustering (defaults to 1:10). |
min_cluster_size |
integer, the minimum number of samples in a cluster (defaults to 5). |
min_cluster_dist |
numeric, the minimum beta difference in cluster centroids (defaults to 0.1). |
max_clusters |
integer, the maximum value of k returned (defaults to 4). |
cores |
integer, specifies the number of cores to use for computation. |
printInfo |
logical, whether to print information during computation or not. |
Betas values should be of type double with samples in
columns and betas in rows. The betas need to be bounded between 0 and 1.
The matrix is typically exported from a GenomicRatioSet,
GenomicMethylSet or MethylSet
object via the getBeta S4 accessor method.
A named vector containing the optimal value for k.
kClusterMethylation, clusterStats
library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)
Harman is a PCA and constrained optimisation based technique that maximises the removal of batch effects from datasets, with the constraint that the probability of overcorrection (i.e. removing genuine biological signal along with batch noise) is kept to a fraction which is set by the end-user (Oytam et al, 2016; http://dx.doi.org/10.1186/s12859-016-1212-5).
Harman expects unbounded data, so for example, with HumanMethylation450 arrays do not use the Beta statistic (with values constrained between 0 and 1), instead use the logit transformed M-values.
harman( datamatrix, expt, batch, limit = 0.95, numrepeats = 100000L, randseed, forceRand = FALSE, printInfo = FALSE )harman( datamatrix, expt, batch, limit = 0.95, numrepeats = 100000L, randseed, forceRand = FALSE, printInfo = FALSE )
datamatrix |
matrix or data.frame, the data values to correct with
samples in columns and data values in rows. Internally, a data.frame will be
coerced to a matrix. Matrices need to be of type |
expt |
vector or factor with the experimental variable of interest (variance to be kept). |
batch |
vector or factor with the batch variable (variance to be removed). |
limit |
numeric, confidence limit. Indicates the limit of confidence in
which to stop removing a batch effect. Must be between |
numrepeats |
integer, the number of repeats in which to run the simulated batch mean distribution estimator using the random selection algorithm. (N.B. 32 bit Windows versions may have an upper limit of 300000 before catastrophic failure) |
randseed |
integer, the seed for random number generation. |
forceRand |
logical, to enforce Harman to use a random selection algorithm to compute corrections. Force the simulated mean code to use random selection of scores to create the simulated batch mean (rather than full explicit calculation from all permutations). |
printInfo |
logical, whether to print information during computation or not. |
The datamatrix needs to be of type integer or
numeric, or alternatively a data.frame that can be coerced into one
using as.matrix. The matrix is to be constructed with data
values (typically microarray probes or sequencing counts) in rows and samples
in columns, much like the 'assayData' slot in the canonical Bioconductor
eSet object, or any object which inherits from it. The data should
have normalisation and any other global adjustment for noise reduction
(such as background correction) applied prior to using Harman.
For converge, the number of simulations, numrepeats parameter should
probably should be at least 100,000. The underlying principle of Harman rests
upon PCA, which is a parametric technique. This implies Harman should be
optimal when the data is normally distributed. However, PCA is known to be
rather robust to very non-normal data.
The output harmanresults object may be presented to summary and data
exploration functions such as plot.harmanresults and
summary.harmanresults as well as the
reconstructData function which creates a corrected matrix of
data with the batch effect removed.
A harmanresults S3 object:
A data.frame of the expt and batch
vectors
The harman runtime parameters. See harman
for details
Confidence intervals and the degree of correction for each principal component
The centering vector returned by prcomp with
center=TRUE
The matrix of eigenvectors (by column) returned from
prcomp
The original PC scores returned by prcomp
The harman corrected PC scores
Oytam et al (2016) BMC Bioinformatics 17:1. DOI: 10.1186/s12859-016-1212-5
reconstructData,
pcaPlot, arrowPlot
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) plot(olf.harman) olf.data.corrected <- reconstructData(olf.harman) ## Reading from a csv file datafile <- system.file("extdata", "NPM_data_first_1000_rows.csv.gz", package="Harman") infofile <- system.file("extdata", "NPM_info.csv.gz", package="Harman") datamatrix <- read.table(datafile, header=TRUE, sep=",", row.names="probeID") batches <- read.table(infofile, header=TRUE, sep=",", row.names="Sample") res <- harman(datamatrix, expt=batches$Treatment, batch=batches$Batch) arrowPlot(res, 1, 3)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) plot(olf.harman) olf.data.corrected <- reconstructData(olf.harman) ## Reading from a csv file datafile <- system.file("extdata", "NPM_data_first_1000_rows.csv.gz", package="Harman") infofile <- system.file("extdata", "NPM_info.csv.gz", package="Harman") datamatrix <- read.table(datafile, header=TRUE, sep=",", row.names="probeID") batches <- read.table(infofile, header=TRUE, sep=",", row.names="Sample") res <- harman(datamatrix, expt=batches$Treatment, batch=batches$Batch) arrowPlot(res, 1, 3)
The S3 object returned after running harman.
harmanresults is the S3 object used to store the results from
harman.
This object may be presented to summary and data exploration functions such
as plot.harmanresults
and summary.harmanresults as well as the
reconstructData function which creates a corrected matrix of
data with the batch effect removed.
factorsA data.frame of the expt and batch
vectors.
parametersThe harman runtime parameters. See harman
for details.
statsConfidence intervals and the degree of correction for each principal component.
centerThe centering vector returned by prcomp with
center=TRUE.
rotationThe matrix of eigenvectors (by column) returned from
prcomp.
originalThe original PC scores returned by prcomp.
correctedThe harman corrected PC scores.
harman, reconstructData,
pcaPlot, arrowPlot
## HarmanResults library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(as.matrix(olf.data), expt, batch) plot(olf.harman) summary(olf.harman) pcaPlot(olf.harman, pc_x=2, pc_y=3) pcaPlot(olf.harman, pc_x=2, pc_y=3, colBy='expt', pch=1) olf.data.corrected <- reconstructData(olf.harman)## HarmanResults library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(as.matrix(olf.data), expt, batch) plot(olf.harman) summary(olf.harman) pcaPlot(olf.harman, pc_x=2, pc_y=3) pcaPlot(olf.harman, pc_x=2, pc_y=3, colBy='expt', pch=1) olf.data.corrected <- reconstructData(olf.harman)
A tweaking of stats::prcomp such that for the svd, the
transpose of u is used instead of v when the number of assays is less than
the number of samples.
harmanScores(x)harmanScores(x)
x |
matrix, data matrix of values to perform PCA on. |
scores, a prcomp-like object with rotation, scores and the center values. The scores are corrected, but all three are needed later to reconstruct the data.
This function is part of a set of three functions to be run in
series. discoverClusteredMethylation takes a matrix of
methylation beta values (typically from the Illumina Infinium Methylation
Assay) and clusters the data across a range of ks specified by the user.
Then the data is reclustered again across the the best two candidate values for k (determined by the rate of change in Bayesian information criterion), and minimum cluster size and distance filters are employed. If both clusters meet these filters, then the higher value of k is returned. This function should be run on uncorrected data that ideally has slides removed which are prone to batch effect. This will bias towards finding clusters that are driven by biological factors such as X-chromosome inactivation and allele-specific methylation.
The output of this function is input for the
kClusterMethylation function which extracts cluster membership
and statistics on variance for a given matrix of beta values. It might be
useful to discover clusters on samples less prone to clustering due to batch
effect or cellular heterogeneity and then recluster all the data for set
values of k via the kClusterMethylation function.
Finally, a comparison of differences of uncorrected to
batch-corrected beta values can be made using clusterStats.
This function generates a data.frame containing log variance ratio and mean
beta differences to clusters after correction.
kClusterMethylation(betas, row_ks, cores = 1, printInfo = FALSE)kClusterMethylation(betas, row_ks, cores = 1, printInfo = FALSE)
betas |
a matrix of methylation beta values with samples in columns and betas in rows. |
row_ks |
vector, the value of k for each row in betas. The names
of |
cores |
integer, specifies the number of cores to use for computation. |
printInfo |
logical, whether to print information during computation or not. |
Betas values should be of type double with samples in
columns and betas in rows. The betas need to be bounded between 0 and 1.
The matrix is typically exported from a GenomicRatioSet,
GenomicMethylSet or MethylSet
object via the getBeta S4 accessor method.
A kClusters S3 object.
discoverClusteredMethylation,
clusterStats
library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)library(HarmanData) data(episcope) bad_batches <- c(1, 5, 9, 17, 25) is_bad_sample <- episcope$pd$array_num %in% bad_batches myK <- discoverClusteredMethylation(episcope$original[, !is_bad_sample]) mykClust = kClusterMethylation(episcope$original, row_ks=myK) res = clusterStats(pre_betas=episcope$original, post_betas=episcope$harman, kClusters = mykClust) all.equal(episcope$ref_md$meandiffs_harman, res$meandiffs) all.equal(episcope$ref_lvr$var_ratio_harman, res$log2_var_ratio)
Generates a Principle Component plot for an instance of
harmanresults.
pcaPlot( harmanresults, pc_x = 1, pc_y = 2, this = "corrected", colBy = "batch", pchBy = "expt", palette = "rainbow", legend = TRUE, col, pch, ... )pcaPlot( harmanresults, pc_x = 1, pc_y = 2, this = "corrected", colBy = "batch", pchBy = "expt", palette = "rainbow", legend = TRUE, col, pch, ... )
harmanresults |
An instance of |
pc_x |
integer, principle component for the plot x dimension. |
pc_y |
integer, principle component for the plot y dimension. |
this |
string, legal values are |
colBy |
string, colour the points by the experimental or batch
variable; legal values
are |
pchBy |
string, point-type by the experimental or batch variable;
legal values are |
palette |
string, the function to call to create a vector of
contiguous colours with the levels of factor in |
legend |
logical, whether to display a legend on the plot. |
col |
colour vector for the points. This parameter overrides
|
pch |
integer vector giving the point type. This parameter
overrides |
... |
further arguments passed to or from other methods. |
If a vector of colours is supplied via the col argument,
then a legend will not be drawn.
None
harmanresults plot.harmanresults
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(as.matrix(olf.data), expt, batch) pcaPlot(olf.harman) pcaPlot(olf.harman, colBy='expt') pcaPlot(olf.harman, pc_x=2, pc_y=3, this='original', pch=17)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(as.matrix(olf.data), expt, batch) pcaPlot(olf.harman) pcaPlot(olf.harman, colBy='expt') pcaPlot(olf.harman, pc_x=2, pc_y=3, this='original', pch=17)
Plot method for instances of harmanresults.
## S3 method for class 'harmanresults' plot(x, ...)## S3 method for class 'harmanresults' plot(x, ...)
x |
An instance of |
... |
further plotting parameters. |
None
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) plot(olf.harman)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) plot(olf.harman)
Generates a Principle Component plot for data.frames, matrices,
or a pre-made prcomp object.
prcompPlot( object, pc_x = 1, pc_y = 2, scale = FALSE, colFactor = NULL, pchFactor = NULL, palette = "rainbow", legend = TRUE, ... )prcompPlot( object, pc_x = 1, pc_y = 2, scale = FALSE, colFactor = NULL, pchFactor = NULL, palette = "rainbow", legend = TRUE, ... )
object |
data.frame, matrix or |
pc_x |
integer, principle component for the plot x dimension. |
pc_y |
integer, principle component for the plot y dimension. |
scale |
logical, whether to scale to unit variance before PCA. |
colFactor |
factor or vector, colour the points by this factor,
default is |
pchFactor |
factor or vector, point-type by this factor,
default is |
palette |
string, the function to call to create a vector of
contiguous colours with |
legend |
logical, whether to display a legend on the plot. |
... |
further arguments passed to or from other methods. |
A data.frame object will be coerced internally to a matrix.
Matrices must be of type double or integer. The
prcompPlot function will then perform a principle component analysis
on the data prior to plotting. The function is call
is prcomp(t(object), retx=TRUE, center=TRUE, scale.=scale).
Instead of specifying a data.frame or matrix, a pre-made prcomp object
can be given to prcompPlot. In this case, care should be taken in
setting the appropriate value of scale.. If a vector is given to
colFactor or pchFactor, they will be coerced internally to
factors.
For the default NULL values of colFactor and pchFactor,
all colours will be black and circles the point type, respectively.
None
library(HarmanData) data(IMR90) expt <- imr90.info$Treatment batch <- imr90.info$Batch prcompPlot(imr90.data, colFactor=expt) pca <- prcomp(t(imr90.data), scale.=TRUE) prcompPlot(pca, 1, 3, colFactor=batch, pchFactor=expt, palette='topo.colors', main='IMR90 PCA plot of Dim 1 and 3')library(HarmanData) data(IMR90) expt <- imr90.info$Treatment batch <- imr90.info$Batch prcompPlot(imr90.data, colFactor=expt) pca <- prcomp(t(imr90.data), scale.=TRUE) prcompPlot(pca, 1, 3, colFactor=batch, pchFactor=expt, palette='topo.colors', main='IMR90 PCA plot of Dim 1 and 3')
Print method for summary.harmanresults.
## S3 method for class 'summary.harmanresults' print(x, ...)## S3 method for class 'summary.harmanresults' print(x, ...)
x |
an object of class |
... |
further parameters. |
Prints summary information from an object of class
summary.harmanresults.
Method which reverts the PCA factorisation for instances of
harmanresults. This allows the original or corrected data to be
returned back from the PCA domain into the original data domain.
reconstructData(object, this = "corrected")reconstructData(object, this = "corrected")
object |
An instance of |
this |
string, legal values are |
matrix of data
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) olf.data.corrected <- reconstructData(olf.harman)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) olf.data.corrected <- reconstructData(olf.harman)
A convienance function for methylation data.
shiftBetas(betas, shiftBy = 1e-04)shiftBetas(betas, shiftBy = 1e-04)
betas |
matrix, beta values. |
shiftBy |
numeric, the amount to shift values of |
None
betas <- seq(0, 1, by=0.05) range(betas) newBetas <- shiftBetas(betas, shiftBy=1e-4) newBetas range(newBetas)betas <- seq(0, 1, by=0.05) range(betas) newBetas <- shiftBetas(betas, shiftBy=1e-4) newBetas range(newBetas)
Summary method for class harmanresults.
## S3 method for class 'harmanresults' summary(object, ...)## S3 method for class 'harmanresults' summary(object, ...)
object |
An object of class |
... |
further parameters. |
Returns an object of class summary.harmanresults.
library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) summary(olf.harman)library(HarmanData) data(OLF) expt <- olf.info$Treatment batch <- olf.info$Batch olf.harman <- harman(olf.data, expt, batch) summary(olf.harman)