Package 'SCBiclust' reference manual

Title:	Identifies Mean, Variance, and Hierarchically Clustered Biclusters
Description:	Identifies a bicluster, a submatrix of the data such that the features and observations within the submatrix differ from those not contained in submatrix, using a two-step method. In the first step, observations in the bicluster are identified to maximize the sum of weighted between cluster feature differences. The method is described in Helgeson et al. (2020) <doi:10.1111/biom.13136>. 'SCBiclust' can be used to identify biclusters which differ based on feature means, feature variances, or more general differences.
Authors:	Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair
Maintainer:	Erika S. Helgeson <[email protected]>
License:	GPL (>= 2)
Version:	1.0.1
Built:	2024-11-20 06:24:09 UTC
Source:	CRAN

'SCBiclust' method for identifying means-based biclusters with Kolmogorov-Smirnov test of feature weights

Description

'SCBiclust' method for identifying means-based biclusters with Kolmogorov-Smirnov test of feature weights

Usage

PermBiclust.beta.ks(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  ks.alpha = 0.05
)
PermBiclust.beta.ks(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  ks.alpha = 0.05
)

Arguments

`x`	a dataset with n rows and p columns, with observations in rows.
`nperms`	number of $Beta(\frac{1}{2}, (p-1)/2)$ distributed variables generated for each feature (default=1000)
`silent`	should progress be printed? (default=TRUE)
`maxnum.bicluster`	The maximum number of biclusters returned
`ks.alpha`	significance level for Kolmogorov-Smirnov test.

Details

Observations in the bicluster are identified such that they maximize the feature-weighted square-root of the between cluster sum of squares. Features in the bicluster are identified based on their contribution to the clustering of the observations. Feature weights are generated in a similar fashion as KMeansSparseCluster except with a modified objective function and no sparsity constraint.

This algoritm uses a numerical approximation to $E(\sqrt{B})$ where $B \sim Beta(\frac{1}{2}, (p-1)/2)$ as the expected null distribution for feature weights. The Kolmogorov-Smirnov test is used to assess if feature weights deviate from the expected null distribution.

Value

The function returns a S3-object with the following attributes:

num.bicluster: The number of biclusters estimated by the procedure.
x.residual: The data matrix x after removing the signals
which.x: A list of length num.bicluster with each list entry containing a logical vector denoting if the data observation is in the given bicluster.
which.y: A list of length num.bicluster with each list entry containing a logical vector denoting if the data feature is in the given bicluster.

Author(s)

Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair

Examples

test <- matrix(rnorm(100*200), nrow=100, ncol=200)
test[1:20,1:20] <- test[1:20,1:20]+rnorm(20*20, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.beta.ks(test, silent=TRUE)
test <- matrix(rnorm(100*200), nrow=100, ncol=200)
test[1:20,1:20] <- test[1:20,1:20]+rnorm(20*20, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.beta.ks(test, silent=TRUE)

'SCBiclust' method for identifying means-based biclusters

Description

'SCBiclust' method for identifying means-based biclusters

Usage

PermBiclust.sigclust(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  alpha = 0.05,
  icovest = 1
)
PermBiclust.sigclust(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  alpha = 0.05,
  icovest = 1
)

Arguments

`x`	a dataset with n rows and p columns, with observations in rows.
`nperms`	number of $Beta(\frac{1}{2}, (p-1)/2)$ distributed variables generated for each feature (default=1000)
`silent`	should progress be printed? (default=TRUE)
`maxnum.bicluster`	The maximum number of biclusters returned
`alpha`	significance level for `sigclust` test.
`icovest`	Coviariance estimation type for `sigclust` test

Details

Observations in the bicluster are identified such that they maximize the feature-weighted between cluster sum of squares. Features in the bicluster are identified based on their contribution to the clustering of the observations. Feature weights are generated in a similar fashion as KMeansSparseCluster except with a modified objective function and no sparsity constraint. This algoritm uses a numerical approximation to $E(\sqrt{B})$ where $B \sim Beta(\frac{1}{2}, (p-1)/2)$ as the expected null distribution for feature weights. The sigclust algorithm is used to test the strength of the identified clusters.

Value

The function returns a S3-object with the following attributes:

num.bicluster: The number of biclusters estimated by the procedure.
x.residual: The data matrix x after removing the signals
which.x: A list of length num.bicluster with each list entry containing a logical vector denoting if the data observation is in the given bicluster.
which.y: A list of length num.bicluster with each list entry containing a logical vector denoting if the data feature is in the given bicluster.

Author(s)

Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair

Examples

 test <- matrix(rnorm(60*180), nrow=60, ncol=180)
test[1:15,1:15] <- test[1:15,1:15]+rnorm(15*15, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.sigclust(test, silent=TRUE)
test <- matrix(rnorm(60*180), nrow=60, ncol=180)
test[1:15,1:15] <- test[1:15,1:15]+rnorm(15*15, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.sigclust(test, silent=TRUE)

'SCBiclust' method for identifying means-based biclusters with optional cluster significance testing

Description

'SCBiclust' method for identifying means-based biclusters with optional cluster significance testing

Usage

PermBiclust.sigclust_stop(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  alpha = 0.05,
  icovest = 1,
  sc = TRUE
)
PermBiclust.sigclust_stop(
  x,
  nperms = 1000,
  silent = TRUE,
  maxnum.bicluster = 5,
  alpha = 0.05,
  icovest = 1,
  sc = TRUE
)

Arguments

`x`	a dataset with n rows and p columns, with observations in rows.
`nperms`	number of $Beta(\frac{1}{2}, (p-1)/2)$ distributed variables generated for each feature (default=1000)
`silent`	should progress be printed? (default=TRUE)
`maxnum.bicluster`	The maximum number of biclusters returned
`alpha`	significance level for `sigclust` test.
`icovest`	Coviariance estimation type for `sigclust` test
`sc`	should the `sigclust` test be used? (default=TRUE)

Details

Observations in the bicluster are identified such that they maximize the feature-weighted between cluster sum of squares. Features in the bicluster are identified based on their contribution to the clustering of the observations. Feature weights are generated in a similar fashion as KMeansSparseCluster except with a modified objective function and no sparsity constraint. This algoritm uses a numerical approximation to $E(\sqrt{B})$ where $B \sim Beta(\frac{1}{2}, (p-1)/2)$ as the expected null distribution for feature weights. Use of the sigclust algorithm to test the strength of the identified clusters is optional in this implementation of the algorithm.

Value

The function returns a S3-object with the following attributes:

num.bicluster: The number of biclusters estimated by the procedure.
x.residual: The data matrix x after removing the signals
which.x: A list of length num.bicluster with each list entry containing a logical vector denoting if the data observation is in the given bicluster.
which.y: A list of length num.bicluster with each list entry containing a logical vector denoting if the data feature is in the given bicluster.

Author(s)

Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair

Examples

test <- matrix(rnorm(60*180), nrow=60, ncol=180)
test[1:15,1:15] <- test[1:15,1:15]+rnorm(15*15, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.sigclust_stop(test, silent=TRUE)
test <- matrix(rnorm(60*180), nrow=60, ncol=180)
test[1:15,1:15] <- test[1:15,1:15]+rnorm(15*15, 2)
test[16:30,51:80] <- test[16:30,51:80]+rnorm(15*30, 3)
PermBiclust.sigclust_stop(test, silent=TRUE)

'SCBiclust' method for identifying hierarchically clustered biclusters

Description

'SCBiclust' method for identifying hierarchically clustered biclusters

Usage

PermHclust.sigclust(
  x = NULL,
  method = c("average", "complete", "single", "centroid"),
  wbound = sqrt(ncol(x)),
  alpha = 0.05,
  dat.perms = 1000,
  dissimilarity = c("squared.distance", "absolute.value"),
  silent = TRUE,
  sigstep = FALSE
)
PermHclust.sigclust(
  x = NULL,
  method = c("average", "complete", "single", "centroid"),
  wbound = sqrt(ncol(x)),
  alpha = 0.05,
  dat.perms = 1000,
  dissimilarity = c("squared.distance", "absolute.value"),
  silent = TRUE,
  sigstep = FALSE
)

Arguments

`x`	a dataset with n rows and p columns, with observations in rows.
`method`	method for agglomeration. See documentation in `hclust`. (default="average")
`wbound`	the tuning parameter for sparse hierarchical clustering. See documentation in `HierarchicalSparseCluster`. (default=`sqrt(ncol(x))`)
`alpha`	significance level for `sigclust` test.
`dat.perms`	number of $Beta(\frac{1}{2}, (p-1)/2)$ distributed variables generated for each feature (default=1000)
`dissimilarity`	How should dissimilarity be calculated? (default is "squared.distance").
`silent`	should progress be printed? (default=TRUE)
`sigstep`	Should `sigclust` be used to assess the strength of identified clusters? (default=FALSE)

Details

Observations in the bicluster are identified such that they maximize the feature-weighted version of the dissimilarity matrix as implemented in HierarchicalSparseCluster. Features in the bicluster are identified based on their contribution to the clustering of the observations. #' This algoritm uses a numerical approximation to $E(\sqrt{B})$ where $B \sim Beta(\frac{1}{2}, (p-1)/2)$ as the expected null distribution for feature weights.

Value

The function returns a S3-object with the following attributes:

which.x: A list of length num.bicluster with each list entry containing a logical vector denoting if the data observation is in the given bicluster.
which.y: A list of length num.bicluster with each list entry containing a logical vector denoting if the data feature is in the given bicluster.

Author(s)

Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair

Examples

test <- matrix(nrow=500, ncol=50)
theta <- rep(NA, 500)
theta[1:300] <- runif(300, 0, pi)
theta[301:500] <- runif(200, pi, 2*pi)
test[1:300,seq(from=2,to=40,by=2)] <- -2+5*sin(theta[1:300])
test[301:500,seq(from=2,to=40,by=2)] <- 5*sin(theta[301:500])
test[1:300,seq(from=1,to=39,by=2)] <- 5+5*cos(theta[1:300])
test[301:500,seq(from=1,to=39,by=2)] <- 5*cos(theta[301:500])
test[,1:40] <- test[,1:40] + rnorm(40*500, 0, 0.2)
test[,41:50] <- rnorm(10*500, 0, 1)
test.PermBiclust <- PermHclust.sigclust(x=test, method='single', dissimilarity='squared.distance')
test <- matrix(nrow=500, ncol=50)
theta <- rep(NA, 500)
theta[1:300] <- runif(300, 0, pi)
theta[301:500] <- runif(200, pi, 2*pi)
test[1:300,seq(from=2,to=40,by=2)] <- -2+5*sin(theta[1:300])
test[301:500,seq(from=2,to=40,by=2)] <- 5*sin(theta[301:500])
test[1:300,seq(from=1,to=39,by=2)] <- 5+5*cos(theta[1:300])
test[301:500,seq(from=1,to=39,by=2)] <- 5*cos(theta[301:500])
test[,1:40] <- test[,1:40] + rnorm(40*500, 0, 0.2)
test[,41:50] <- rnorm(10*500, 0, 1)
test.PermBiclust <- PermHclust.sigclust(x=test, method='single', dissimilarity='squared.distance')

'SCBiclust' method for identifying variance-based biclusters

Description

'SCBiclust' method for identifying variance-based biclusters

Usage

VarPermBiclust.chisqdiff(
  x,
  min.size = max(5, round(nrow(x)/20)),
  nperms = 1000,
  silent = TRUE
)
VarPermBiclust.chisqdiff(
  x,
  min.size = max(5, round(nrow(x)/20)),
  nperms = 1000,
  silent = TRUE
)

Arguments

`x`	a dataset with n rows and p columns, with observations in rows.
`min.size`	Minimum size of observations included in a valid bicluster (default=`max(5,round(nrow(x)/20)))`
`nperms`	number of $\chi^2_{n_1}$ and $\chi^2_{n_2}$ variables generated for each feature where $n_1$ and $n_2$ are the number of observations in cluster 1 and cluster 2, respectively. (default=100)
`silent`	should progress be printed? (default=TRUE)

Details

Observations in the bicluster are identified such that they maximize the feature-weighted sum of between cluster difference in feature variances. Features in the bicluster are identified based on their contribution to the clustering of the observations. This algoritm uses a numerical approximation $log(abs(\chi^2_{n_1}-chi^2_{n_2})+1)$ as the expected null distribution for feature weights.

VarPermBiclust.chisqdiff will identify at most one variance bicluster. To identify additional biclusters first the feature signal of the identified bicluster should be removed by scaling the variance of elements in the previously identified bicluster, Then VarPermBiclust.chisqdiff can be used on the residual data matrix. (see example)

Value

The function returns a S3-object with the following attributes:

which.x: A list of length num.bicluster with each list entry containing a logical vector denoting if the data observation is in the given bicluster.
which.y: A list of length num.bicluster with each list entry containing a logical vector denoting if the data feature is in the given bicluster.

Author(s)

Erika S. Helgeson, Qian Liu, Guanhua Chen, Michael R. Kosorok , and Eric Bair

Examples

test <- matrix(rnorm(100*50, mean=1, sd=2), nrow=100)
test[1:30, 1:20] <- matrix(rnorm(30*20, mean=1, sd=15), nrow=30)
test.VarPermBiclust <- VarPermBiclust.chisqdiff(test)
x=test.VarPermBiclust$which.x
y=test.VarPermBiclust$which.y
# Code for identifying additional biclusters after removing bicluster signal

temp <- scale(test)
temp[x,y] <-t(t(temp[x,y])*(apply(temp[!x,y],2,sd)/
                              apply(temp[x,y],2,sd)))
test.VarPermBiclust.2 <- VarPermBiclust.chisqdiff(temp)

test <- matrix(rnorm(100*50, mean=1, sd=2), nrow=100)
test[1:30, 1:20] <- matrix(rnorm(30*20, mean=1, sd=15), nrow=30)
test.VarPermBiclust <- VarPermBiclust.chisqdiff(test)
x=test.VarPermBiclust$which.x
y=test.VarPermBiclust$which.y
# Code for identifying additional biclusters after removing bicluster signal

temp <- scale(test)
temp[x,y] <-t(t(temp[x,y])*(apply(temp[!x,y],2,sd)/
                              apply(temp[x,y],2,sd)))
test.VarPermBiclust.2 <- VarPermBiclust.chisqdiff(temp)

Package 'SCBiclust'

Help Index

'SCBiclust' method for identifying means-based biclusters with Kolmogorov-Smirnov test of feature weights

Description

Usage

Arguments

Details

Value

Author(s)

Examples

'SCBiclust' method for identifying means-based biclusters

Description

Usage

Arguments

Details

Value

Author(s)

Examples

'SCBiclust' method for identifying means-based biclusters with optional cluster significance testing

Description

Usage

Arguments

Details

Value

Author(s)

Examples

'SCBiclust' method for identifying hierarchically clustered biclusters

Description

Usage

Arguments

Details

Value

Author(s)

Examples

'SCBiclust' method for identifying variance-based biclusters

Description

Usage

Arguments

Details

Value

Author(s)

Examples