Package 'CoGAPS'

Title: Coordinated Gene Activity in Pattern Sets
Description: Coordinated Gene Activity in Pattern Sets (CoGAPS) implements a Bayesian MCMC matrix factorization algorithm, GAPS, and links it to gene set statistic methods to infer biological process activity. It can be used to perform sparse matrix factorization on any data, and when this data represents biomolecules, to do gene set analysis.
Authors: Jeanette Johnson, Ashley Tsang, Jacob Mitchell, Thomas Sherman, Wai-shing Lee, Conor Kelton, Ondrej Maxian, Jacob Carey, Genevieve Stein-O'Brien, Michael Considine, Maggie Wodicka, John Stansfield, Shawn Sivy, Carlo Colantuoni, Alexander Favorov, Mike Ochs, Elana Fertig
Maintainer: Elana J. Fertig <[email protected]>, Thomas D. Sherman <[email protected]>, Jeanette Johnson <[email protected]>, Dmitrijs Lvovs <[email protected]>
License: BSD_3_clause + file LICENSE
Version: 3.25.1
Built: 2024-07-18 02:59:53 UTC
Source: https://github.com/bioc/CoGAPS

Help Index

CoGAPS: Coordinated Gene Activity in Pattern Sets

Description

CoGAPS implements a Bayesian MCMC matrix factorization algorithm, GAPS, and links it to gene set statistic methods to infer biological process activity. It can be used to perform sparse matrix factorization on any data, and when this data represents biomolecules, to do gene set analysis.

Package: CoGAPS
Type: Package
Version: 2.99.0
Date: 2018-01-24
License: LGPL

Author(s)

Maintainer: Elana J. Fertig [email protected], Michael F. Ochs [email protected]

References

Fertig EJ, Ding J, Favorov AV, Parmigiani G, Ochs MF. CoGAPS: an R/C++ package to identify patterns and biological process activity in transcriptomic data. Bioinformatics. 2010 Nov 1;26(21):2792-3

binary heatmap for standardized feature matrix

Description

creates a binarized heatmap of the A matrix in which the value is 1 if the value in Amean is greater than threshold * Asd and 0 otherwise

Usage

binaryA(object, threshold = 3)

## S4 method for signature 'CogapsResult'
binaryA(object, threshold = 3)

Arguments

object

an object of type CogapsResult
threshold

the number of standard deviations above zero that an element of Amean must be to get a value of 1

Value

plots a heatmap of the A Matrix

Examples

data(GIST)
# to expensive to call since it plots
# binaryA(GIST.result, threshold=3)

Information About Package Compilation

Description

Information About Package Compilation

Usage

buildReport()

Details

returns information about how the package was compiled, i.e. which compiler/version was used, which compile time options were enabled, etc...

Value

string containing build report

Examples

CoGAPS::buildReport()

calculate statistic on sets of measurements (genes) or samples

Description

calculates a statistic to determine if a pattern is enriched in a a particular set of measurements or samples.

Usage

calcCoGAPSStat(
  object,
  sets = NULL,
  whichMatrix = "featureLoadings",
  numPerm = 1000,
  ...
)

## S4 method for signature 'CogapsResult'
calcCoGAPSStat(
  object,
  sets = NULL,
  whichMatrix = "featureLoadings",
  numPerm = 1000,
  ...
)

Arguments

object

an object of type CogapsResult
sets

list of sets of measurements/samples
whichMatrix

either "featureLoadings" or "sampleFactors" indicating which matrix to calculate the statistics for
numPerm

number of permutations to use when calculatin p-value
...

handles old arguments for backwards compatibility

Value

gene set statistics for each column of A

probability gene belongs in gene set

Description

calculates the probability that a gene listed in a gene set behaves like other genes in the set within the given data set

Usage

calcGeneGSStat(
  object,
  GStoGenes,
  numPerm,
  Pw = rep(1, ncol(object@featureLoadings)),
  nullGenes = FALSE
)

## S4 method for signature 'CogapsResult'
calcGeneGSStat(
  object,
  GStoGenes,
  numPerm,
  Pw = rep(1, ncol(object@featureLoadings)),
  nullGenes = FALSE
)

Arguments

object

an object of type CogapsResult
GStoGenes

data.frame or list with gene sets
numPerm

number of permutations for null
Pw

weight on genes
nullGenes

logical indicating gene adjustment

Value

gene similiarity statistic

compute z-score matrix

Description

calculates the Z-score for each element based on input mean and standard deviation matrices

Usage

calcZ(object, whichMatrix)

## S4 method for signature 'CogapsResult'
calcZ(object, whichMatrix)

Arguments

object

an object of type CogapsResult
whichMatrix

either "featureLoadings" or "sampleFactors" indicating which matrix to calculate the z-score for

Value

matrix of z-scores

Examples

data(GIST)
featureZScore <- calcZ(GIST.result, "featureLoadings")

Check if package was built with checkpoints enabled

Description

Check if package was built with checkpoints enabled

Usage

checkpointsEnabled()

Value

true/false if checkpoints are enabled

Examples

CoGAPS::checkpointsEnabled()

CoGAPS Matrix Factorization Algorithm

Description

calls the C++ MCMC code and performs Bayesian matrix factorization returning the two matrices that reconstruct the data matrix

Usage

CoGAPS(
  data,
  params = new("CogapsParams"),
  nThreads = 1,
  messages = TRUE,
  outputFrequency = 1000,
  uncertainty = NULL,
  checkpointOutFile = "gaps_checkpoint.out",
  checkpointInterval = 0,
  checkpointInFile = NULL,
  transposeData = FALSE,
  BPPARAM = NULL,
  workerID = 1,
  asynchronousUpdates = TRUE,
  nSnapshots = 0,
  snapshotPhase = "sampling",
  ...
)

Arguments

data

File name or R object (see details for supported types)
params

CogapsParams object
nThreads

maximum number of threads to run on
messages

T/F for displaying output
outputFrequency

number of iterations between each output (set to 0 to disable status updates, other output is controlled by @code messages)
uncertainty

uncertainty matrix - either a matrix or a supported file type
checkpointOutFile

name of the checkpoint file to create
checkpointInterval

number of iterations between each checkpoint (set to 0 to disable checkpoints)
checkpointInFile

if this is provided, CoGAPS runs from the checkpoint contained in this file
transposeData

T/F for transposing data while reading it in - useful for data that is stored as samples x genes since CoGAPS requires data to be genes x samples
BPPARAM

BiocParallel backend
workerID

if calling CoGAPS in parallel the worker ID can be specified, only worker 1 prints output and each worker outputs when it finishes, this is not neccesary when using the default parallel methods (i.e. distributed CoGAPS) but only when the user is manually calling CoGAPS in parallel
asynchronousUpdates

enable asynchronous updating which allows for multi-threaded runs
nSnapshots

how many snapshots to take in each phase, setting this to 0 disables snapshots
snapshotPhase

which phase to take snapsjots in e.g. "equilibration", "sampling", "all"
...

allows for overwriting parameters in params

Details

The supported R types are: matrix, data.frame, SummarizedExperiment, SingleCellExperiment. The supported file types are csv, tsv, and mtx.

Value

CogapsResult object

Examples

# Running from R object
data(GIST)
resultA <- CoGAPS(GIST.data_frame, nIterations=25)

# Running from file name
gist_path <- system.file("extdata/GIST.mtx", package="CoGAPS")
resultB <- CoGAPS(gist_path, nIterations=25)

# Setting Parameters
params <- new("CogapsParams")
params <- setParam(params, "nPatterns", 3)
resultC <- CoGAPS(GIST.data_frame, params, nIterations=25)

CogapsParams constructor

Description

create a CogapsParams object

Usage

CogapsParams(...)

Arguments

...

parameters for the initialization method

Value

CogapsParams object

Examples

params <- CogapsParams(nPatterns=10)
params

CogapsParams

Description

Encapsulates all parameters for the CoGAPS algorithm

Slots

nPatterns

number of patterns CoGAPS will learn

nIterations

number of iterations for each phase of the algorithm

alphaA

sparsity parameter for feature matrix

alphaP

sparsity parameter for sample matrix

maxGibbsMassA

atomic mass restriction for feature matrix

maxGibbsMassP

atomic mass restriction for sample matrix

seed

random number generator seed

sparseOptimization

speeds up performance with sparse data (roughly >80 default uncertainty

distributed

either "genome-wide" or "single-cell" indicating which distributed algorithm should be used

nSets

[distributed parameter] number of sets to break data into

cut

[distributed parameter] number of branches at which to cut dendrogram used in pattern matching

minNS

[distributed parameter] minimum of individual set contributions a cluster must contain

maxNS

[distributed parameter] maximum of individual set contributions a cluster can contain

explicitSets

[distributed parameter] specify subsets by index or name

samplingAnnotation

[distributed parameter] specify categories along the rows (cols) to use for weighted sampling

samplingWeight

[distributed parameter] weights associated with samplingAnnotation

subsetIndices

set of indices to use from the data

subsetDim

which dimension (1=rows, 2=cols) to subset

geneNames

vector of names of genes in data

sampleNames

vector of names of samples in data

fixedPatterns

fix either 'A' or 'P' matrix to these values, in the context of distributed CoGAPS (GWCoGAPS/scCoGAPS), the first phase is skipped and fixedPatterns is used for all sets - allowing manual pattern matching, as well as fixed runs of standard CoGAPS

whichMatrixFixed

either 'A' or 'P', indicating which matrix is fixed

takePumpSamples

whether or not to take PUMP samples

checkpointInterval

how many iterations between each checkpoint (set to 0 to disable)

checkpointInFile

file path to load checkpoint from

checkpointOutFile

file path where checkpoint should be written to

CogapsResult

Description

Contains all output from Cogaps run

Slots

factorStdDev

std dev of the sampled P matrices

loadingStdDev

std dev of the sampled A matrices

Check if compiler supported OpenMP

Description

Check if compiler supported OpenMP

Usage

compiledWithOpenMPSupport()

Value

true/false if OpenMP was supported

Examples

CoGAPS::compiledWithOpenMPSupport()

compute gene probability

Description

Computes the p-value for gene set membership using the CoGAPS-based statistics developed in Fertig et al. (2012). This statistic refines set membership for each candidate gene in a set specified in GSGenes by comparing the inferred activity of that gene to the average activity of the set.

Usage

computeGeneGSProb(
  object,
  GStoGenes,
  numPerm = 500,
  Pw = rep(1, ncol(object@featureLoadings)),
  PwNull = FALSE
)

## S4 method for signature 'CogapsResult'
computeGeneGSProb(
  object,
  GStoGenes,
  numPerm = 500,
  Pw = rep(1, ncol(object@featureLoadings)),
  PwNull = FALSE
)

Arguments

object

an object of type CogapsResult
GStoGenes

data.frame or list with gene sets
numPerm

number of permutations for null
Pw

weight on genes
PwNull

- logical indicating gene adjustment

Value

A vector of length GSGenes containing the p-values of set membership for each gene containined in the set specified in GSGenes.

find the consensus pattern matrix across all subsets

Description

find the consensus pattern matrix across all subsets

Usage

findConsensusMatrix(unmatchedPatterns, gapsParams)

Arguments

unmatchedPatterns

list of all unmatched pattern matrices from initial run of CoGAPS
gapsParams

list of all CoGAPS parameters

Value

matrix of consensus patterns

read CoGAPS Result object from a directory with a set of csvs see toCSV

Description

save as csv

Usage

fromCSV(save_location = ".")

## S4 method for signature 'character'
fromCSV(save_location = ".")

Arguments

save_location

directory to read from

Value

CogapsResult object

return Amplitude matrix from CogapsResult object

Description

return Amplitude matrix from CogapsResult object

Usage

getAmplitudeMatrix(object)

## S4 method for signature 'CogapsResult'
getAmplitudeMatrix(object)

Arguments

object

an object of type CogapsResult

Value

amplitude matrix

Examples

data(GIST)
amplitudeMatrix <- getAmplitudeMatrix(GIST.result)

return clustered patterns from set of all patterns across all subsets

Description

return clustered patterns from set of all patterns across all subsets

Usage

getClusteredPatterns(object)

## S4 method for signature 'CogapsResult'
getClusteredPatterns(object)

Arguments

object

an object of type CogapsResult

Value

CogapsParams object

Examples

data(GIST)
clusteredPatterns <- getClusteredPatterns(GIST.result)

return correlation between each pattern and the cluster mean

Description

return correlation between each pattern and the cluster mean

Usage

getCorrelationToMeanPattern(object)

## S4 method for signature 'CogapsResult'
getCorrelationToMeanPattern(object)

Arguments

object

an object of type CogapsResult

Value

CogapsParams object

Examples

data(GIST)
corrToMeanPattern <- getCorrelationToMeanPattern(GIST.result)

return featureLoadings matrix from CogapsResult object

Description

return featureLoadings matrix from CogapsResult object

Usage

getFeatureLoadings(object)

## S4 method for signature 'CogapsResult'
getFeatureLoadings(object)

Arguments

object

an object of type CogapsResult

Value

featureLoadings matrix

Examples

data(GIST)
fLoadings <- getFeatureLoadings(GIST.result)

return chi-sq of final matrices

Description

return chi-sq of final matrices

Usage

getMeanChiSq(object)

## S4 method for signature 'CogapsResult'
getMeanChiSq(object)

Arguments

object

an object of type CogapsResult

Value

chi-sq error

Examples

data(GIST)
getMeanChiSq(GIST.result)

return original parameters used to generate this result

Description

return original parameters used to generate this result

Usage

getOriginalParameters(object)

## S4 method for signature 'CogapsResult'
getOriginalParameters(object)

Arguments

object

an object of type CogapsResult

Value

CogapsParams object

Examples

data(GIST)
params <- getOriginalParameters(GIST.result)

get the value of a parameter

Description

get the value of a parameter

Usage

getParam(object, whichParam)

## S4 method for signature 'CogapsParams'
getParam(object, whichParam)

Arguments

object

an object of type CogapsParams
whichParam

a string with the name of the requested parameter

Value

the value of the parameter

Examples

params <- new("CogapsParams")
 getParam(params, "seed")

generate statistics associating patterns with gene sets

Description

generate statistics associating patterns with gene sets

Usage

getPatternGeneSet(
  object,
  gene.sets,
  method = c("enrichment", "overrepresentation"),
  ...
)

## S4 method for signature 'CogapsResult,list,character'
getPatternGeneSet(
  object,
  gene.sets,
  method = c("enrichment", "overrepresentation"),
  ...
)

Arguments

object

an object of type CogapsResult
gene.sets

a list of gene sets to test. List names should be the names of the gene sets
method

enrichment or overrepresentation. Conducts a test for gene set enrichment using fgsea::gsea ranking features by pattern amplitude or a test for gene set overrepresentation in pattern markers using fgsea::fora, respectively.
...

additional parameters passed to patternMarkers if using overrepresentation method

Value

list of dataframes containing gene set enrichment or gene set overrepresentation statistics

Examples

data(GIST)
gs.test <- list(
"gs1" = c("Hs.2", "Hs.4", "Hs.36", "Hs.96", "Hs.202"),
"gs2" = c("Hs.699463", "Hs.699288", "Hs.699280", "Hs.699154", "Hs.697294")
)
getPatternGeneSet(object = GIST.result, gene.sets = gs.test, method = "enrichment")
getPatternGeneSet(object = GIST.result, gene.sets = gs.test, method = "overrepresentation")

return pattern matrix from CogapsResult object

Description

return pattern matrix from CogapsResult object

Usage

getPatternMatrix(object)

## S4 method for signature 'CogapsResult'
getPatternMatrix(object)

Arguments

object

an object of type CogapsResult

Value

pattern matrix

Examples

data(GIST)
patternMatrix <- getPatternMatrix(GIST.result)

get specified number of retina subsets

Description

combines retina subsets from extdata directory

Usage

getRetinaSubset(n = 1)

Arguments

n

number of subsets to use

Value

matrix of RNA counts

Examples

retSubset <- getRetinaSubset()
dim(retSubset)

return sampleFactors matrix from CogapsResult object

Description

return sampleFactors matrix from CogapsResult object

Usage

getSampleFactors(object)

## S4 method for signature 'CogapsResult'
getSampleFactors(object)

Arguments

object

an object of type CogapsResult

Value

sampleFactors matrix

Examples

data(GIST)
sFactors <- getSampleFactors(GIST.result)

return the names of the genes (samples) in each subset

Description

return the names of the genes (samples) in each subset

Usage

getSubsets(object)

## S4 method for signature 'CogapsResult'
getSubsets(object)

Arguments

object

an object of type CogapsResult

Value

CogapsParams object

Examples

data(GIST)
subsets <- getSubsets(GIST.result)

return unmatched patterns from each subset

Description

return unmatched patterns from each subset

Usage

getUnmatchedPatterns(object)

## S4 method for signature 'CogapsResult'
getUnmatchedPatterns(object)

Arguments

object

an object of type CogapsResult

Value

CogapsParams object

Examples

data(GIST)
unmatchedPatterns <- getUnmatchedPatterns(GIST.result)

return version number used to generate this result

Description

return version number used to generate this result

Usage

getVersion(object)

## S4 method for signature 'CogapsResult'
getVersion(object)

Arguments

object

an object of type CogapsResult

Value

version number

Examples

data(GIST)
getVersion(GIST.result)

GIST gene expression data from Ochs et al. (2009)

Description

GIST gene expression data from Ochs et al. (2009)

GIST gene expression data from Ochs et al. (2009)

Description

GIST gene expression data from Ochs et al. (2009)

CoGAPS result from running on GIST dataset

Description

CoGAPS result from running on GIST dataset

GIST gene expression uncertainty matrix from Ochs et al. (2009)

Description

GIST gene expression uncertainty matrix from Ochs et al. (2009)

Genome Wide CoGAPS

Description

wrapper around genome-wide distributed algorithm for CoGAPS

Usage

GWCoGAPS(
  data,
  params = new("CogapsParams"),
  nThreads = 1,
  messages = TRUE,
  outputFrequency = 500,
  uncertainty = NULL,
  checkpointOutFile = "gaps_checkpoint.out",
  checkpointInterval = 1000,
  checkpointInFile = NULL,
  transposeData = FALSE,
  BPPARAM = NULL,
  workerID = 1,
  asynchronousUpdates = FALSE,
  ...
)

Arguments

data

File name or R object (see details for supported types)
params

CogapsParams object
nThreads

maximum number of threads to run on
messages

T/F for displaying output
outputFrequency

number of iterations between each output (set to 0 to disable status updates, other output is controlled by @code messages)
uncertainty

uncertainty matrix - either a matrix or a supported file type
checkpointOutFile

name of the checkpoint file to create
checkpointInterval

number of iterations between each checkpoint (set to 0 to disable checkpoints)
checkpointInFile

if this is provided, CoGAPS runs from the checkpoint contained in this file
transposeData

T/F for transposing data while reading it in - useful for data that is stored as samples x genes since CoGAPS requires data to be genes x samples
BPPARAM

BiocParallel backend
workerID

if calling CoGAPS in parallel the worker ID can be specified, only worker 1 prints output and each worker outputs when it finishes, this is not neccesary when using the default parallel methods (i.e. distributed CoGAPS) but only when the user is manually calling CoGAPS in parallel
asynchronousUpdates

enable asynchronous updating which allows for multi-threaded runs
...

allows for overwriting parameters in params

Value

CogapsResult object

Examples

## Not run: 
data(GIST)
params <- new("CogapsParams")
params <- setDistributedParams(params, nSets=2)
params <- setParam(params, "nIterations", 100)
params <- setParam(params, "nPatterns", 3)
result <- GWCoGAPS(GIST.matrix, params, BPPARAM=BiocParallel::SerialParam())

## End(Not run)

constructor for CogapsParams

Description

constructor for CogapsParams

Usage

## S4 method for signature 'CogapsParams'
initialize(.Object, distributed = NULL, ...)

Arguments

.Object

CogapsParams object
distributed

either "genome-wide" or "single-cell" indicating which distributed algorithm should be used
...

initial values for slots

Value

initialized CogapsParams object

Constructor for CogapsResult

Description

Constructor for CogapsResult

Usage

## S4 method for signature 'CogapsResult'
initialize(
  .Object,
  Amean,
  Pmean,
  Asd,
  Psd,
  meanChiSq,
  geneNames,
  sampleNames,
  diagnostics = NULL,
  ...
)

Arguments

.Object

CogapsResult object
Amean

mean of sampled A matrices
Pmean

mean of sampled P matrices
Asd

std dev of sampled A matrices
Psd

std dev of sampled P matrices
meanChiSq

mean value of ChiSq statistic
geneNames

names of genes in data
sampleNames

names of samples in data
diagnostics

assorted diagnostic reports from the run
...

initial values for slots

Value

initialized CogapsResult object

MANOVA statistical test for patterns between sample groups

Description

MANOVA statistical test–wraps base R manova

Usage

MANOVA(interestedVariables, object)

## S4 method for signature 'matrix,CogapsResult'
MANOVA(interestedVariables, object)

Arguments

interestedVariables

study design for manova
object

CogapsResult object

Value

list of manova fit results

Toy example to run CoGAPS on.

Description

  • V1..V20. some variables, for example levels of gene expression

Usage

data(modsimdata)

Format

'data.frame': 25 obs. of 20 variables.

Result of applying CoGAPS on the Toy example.

Description

Result of applying CoGAPS on the Toy example.

Usage

data(modsimresult)

Format

S4 class ‘CogapsResult’ [package “CoGAPS”] with 7 slots.

compute pattern markers statistic

Description

estimate the most associated pattern for each feature

Usage

patternMarkers(object, threshold = "all", lp = NULL, axis = 1)

## S4 method for signature 'CogapsResult'
patternMarkers(object, threshold = "all", lp = NULL, axis = 1)

Arguments

object

an object of type CogapsResult
threshold

the type of threshold to be used. The default "all" will distribute features into patterns with the highest ranking as ranked by the increasing Euclidian distance between feature loadings and lp. The alternative "cut" will only keep the features that are ranked higher than the first feature having greater intra-pattern compared to inter-pattern rank. This is useful to limit the number of markers ranked similarly everywhere. Features may be present in multiple patterns for "cut".
lp

list of vectors of weights for each pattern to be used for finding markers. If NULL, list of synthetic one-hot markers for each pattern will be generated and matched against.
axis

controls the matrix to use for ranking. 1 for featureLoadings, 2 for sampleFactors.

Value

List of: list of marker features for each pattern (best rank first), a matrix of ranks of each feature in each pattern, a matrix of scores for each feature in each pattern.

List of: list of marker features for each pattern (best rank first), and a matrix of ranks of each feature in each pattern.

Examples

data(GIST)
pm <- patternMarkers(GIST.result)

generate a barchart of most significant hallmark sets for a pattern

Description

generate a barchart of most significant hallmark sets for a pattern

Usage

plotPatternGeneSet(patterngeneset, whichpattern = 1, padj_threshold = 0.05)

## S4 method for signature 'list,numeric,numeric'
plotPatternGeneSet(patterngeneset, whichpattern = 1, padj_threshold = 0.05)

Arguments

patterngeneset

output from getPatternGeneSet
whichpattern

which pattern to generate bar chart for
padj_threshold

maximum adjusted p-value of gene sets rendered on the resulting plot

Value

image object of barchart

heatmap of original data clustered by pattern markers statistic

Description

heatmap of original data clustered by pattern markers statistic

Usage

plotPatternMarkers(
  object,
  data,
  patternMarkers,
  patternPalette,
  sampleNames,
  samplePalette = NULL,
  heatmapCol = bluered,
  colDendrogram = TRUE,
  scale = "row",
  ...
)

Arguments

object

an object of type CogapsResult
data

the original data as a matrix
patternMarkers

pattern markers to be plotted, as generated by the patternMarkers function
patternPalette

a vector indicating what color should be used for each pattern
sampleNames

names of the samples to use for labeling
samplePalette

a vector indicating what color should be used for each sample
heatmapCol

pallelet giving color scheme for heatmap
colDendrogram

logical indicating whether to display sample dendrogram
scale

character indicating if the values should be centered and scaled in either the row direction or the column direction, or none. The default is "row".
...

additional graphical parameters to be passed to heatmap.2

Value

heatmap of the data values for the patternMarkers

See Also

heatmap.2

plot of residuals

Description

calculate residuals and produce heatmap

Usage

plotResiduals(object, data, uncertainty = NULL)

## S4 method for signature 'CogapsResult'
plotResiduals(object, data, uncertainty = NULL)

Arguments

object

an object of type CogapsResult
data

original data matrix run through GAPS
uncertainty

original standard deviation matrix run through GAPS

Value

creates a residual plot

Examples

data(GIST)
# to expensive to call since it plots
# plotResiduals(GIST.result, GIST.matrix)

reconstruct gene

Description

reconstruct gene

Usage

reconstructGene(object, genes = NULL)

## S4 method for signature 'CogapsResult'
reconstructGene(object, genes = NULL)

Arguments

object

an object of type CogapsResult
genes

an index of the gene or genes of interest

Value

the D' estimate of a gene or set of genes

Examples

data(GIST)
estimatedD <- reconstructGene(GIST.result)

Single Cell CoGAPS

Description

wrapper around single-cell distributed algorithm for CoGAPS

Usage

scCoGAPS(
  data,
  params = new("CogapsParams"),
  nThreads = 1,
  messages = TRUE,
  outputFrequency = 500,
  uncertainty = NULL,
  checkpointOutFile = "gaps_checkpoint.out",
  checkpointInterval = 1000,
  checkpointInFile = NULL,
  transposeData = FALSE,
  BPPARAM = NULL,
  workerID = 1,
  asynchronousUpdates = FALSE,
  ...
)

Arguments

data

File name or R object (see details for supported types)
params

CogapsParams object
nThreads

maximum number of threads to run on
messages

T/F for displaying output
outputFrequency

number of iterations between each output (set to 0 to disable status updates, other output is controlled by @code messages)
uncertainty

uncertainty matrix - either a matrix or a supported file type
checkpointOutFile

name of the checkpoint file to create
checkpointInterval

number of iterations between each checkpoint (set to 0 to disable checkpoints)
checkpointInFile

if this is provided, CoGAPS runs from the checkpoint contained in this file
transposeData

T/F for transposing data while reading it in - useful for data that is stored as samples x genes since CoGAPS requires data to be genes x samples
BPPARAM

BiocParallel backend
workerID

if calling CoGAPS in parallel the worker ID can be specified, only worker 1 prints output and each worker outputs when it finishes, this is not neccesary when using the default parallel methods (i.e. distributed CoGAPS) but only when the user is manually calling CoGAPS in parallel
asynchronousUpdates

enable asynchronous updating which allows for multi-threaded runs
...

allows for overwriting parameters in params

Value

CogapsResult object

Examples

## Not run: 
data(GIST)
params <- new("CogapsParams")
params <- setDistributedParams(params, nSets=2)
params <- setParam(params, "nIterations", 100)
params <- setParam(params, "nPatterns", 3)
result <- scCoGAPS(t(GIST.matrix), params, BPPARAM=BiocParallel::SerialParam())

## End(Not run)

set the annotation labels and weights for subsetting the data

Description

these parameters are interrelated so they must be set together

Usage

setAnnotationWeights(object, annotation, weights)

## S4 method for signature 'CogapsParams'
setAnnotationWeights(object, annotation, weights)

Arguments

object

an object of type CogapsParams
annotation

vector of labels
weights

vector of weights

Value

the modified params object

Examples

params <- new("CogapsParams")
 params <- setAnnotationWeights(params, c('a', 'b', 'c'), c(1,2,1))

set the value of parameters for distributed CoGAPS

Description

these parameters are interrelated so they must be set together

Usage

setDistributedParams(
  object,
  nSets = NULL,
  cut = NULL,
  minNS = NULL,
  maxNS = NULL
)

## S4 method for signature 'CogapsParams'
setDistributedParams(
  object,
  nSets = NULL,
  cut = NULL,
  minNS = NULL,
  maxNS = NULL
)

Arguments

object

an object of type CogapsParams
nSets

number of sets to break data into
cut

number of branches at which to cut dendrogram used in pattern matching
minNS

minimum of individual set contributions a cluster must contain
maxNS

maximum of individual set contributions a cluster can contain

Value

the modified params object

Examples

params <- new("CogapsParams")
 params <- setDistributedParams(params, 5)

set the fixed patterns for either the A or the P matrix

Description

these parameters are interrelated so they must be set together

Usage

setFixedPatterns(object, fixedPatterns, whichMatrixFixed)

## S4 method for signature 'CogapsParams'
setFixedPatterns(object, fixedPatterns, whichMatrixFixed)

Arguments

object

an object of type CogapsParams
fixedPatterns

values for either the A or P matrix
whichMatrixFixed

either 'A' or 'P' indicating which matrix is fixed

Value

the modified params object

Examples

params <- new("CogapsParams")
data(GIST)
params <- setFixedPatterns(params, getSampleFactors(GIST.result), 'P')

set the value of a parameter

Description

set the value of a parameter

Usage

setParam(object, whichParam, value)

## S4 method for signature 'CogapsParams'
setParam(object, whichParam, value)

Arguments

object

an object of type CogapsParams
whichParam

a string with the name of the parameter to be changed
value

the value to set the parameter to

Value

the modified params object

Examples

params <- new("CogapsParams")
 params <- setParam(params, "seed", 123)

save CoGAPS Result object as a set of csvs to directory see fromCSV

Description

save as csv

Usage

toCSV(object, save_location = ".")

## S4 method for signature 'CogapsResult,character'
toCSV(object, save_location = ".")

Arguments

object

CogapsResult object
save_location

directory to write to

Value

none