Package 'CBN2Path'

Title: CBN2Path: an R/Bioconductor package for the analysis of cancer progression pathways using Conjunctive Bayesian Networks
Description: CBN2Path package provides a unifying interface to facilitate CBN-based quantification, analysis and visualization of cancer progression pathways.
Authors: William Choi-Kim [aut, cre], Sayed-Rzgar Hosseini [aut, cre]
Maintainer: William Choi-Kim <[email protected]>, Sayed-Rzgar Hosseini <[email protected]>
License: MIT + file LICENSE
Version: 1.3.0
Built: 2026-05-29 08:31:40 UTC
Source: https://github.com/bioc/CBN2Path

Help Index

base2Indexing

Description

base2Indexing

Usage

base2Indexing(mat)

Arguments

mat

A given poset represented by a binary matrix (in B-CBN)

Value

#Poset weight vectors based on the frequency of occurrence in the BCBN MCMC-sampling scheme.

Examples

set.seed(100)
mat <- matrix(sample(c(0, 1), 16, replace = TRUE), 4, 4)
base2Indexing(mat)

base2IndVec

Description

base2IndVec

Usage

base2IndVec(vec)

Arguments

vec

a binary genotype vector

Value

a number used for indexing a given genotype

Examples

vec <- c(0, 1, 0, 1)
base2IndVec(vec)

B-CBN

Description

B-CBN

Usage

bcbn(
  data = defaultData(),
  nSamples = 25000,
  theta = 0,
  epsilon = 0.05,
  nChains = 4,
  thin = 10,
  maxL = 1000,
  nCores = 1,
  progressBar = FALSE
)

Arguments

data

Generated data
nSamples

Number of samples <def: 25000>
theta

Theta <def: 0>
epsilon

Epsilon <def: 0.05>
nChains

N-Chains <def: 4>
thin

Thin <def: 10>
maxL

The maximum number of iteration <def: 1000>
nCores

Number of parallelized cores <def: 1>
progressBar

Print out progress bar; default is FALSE

Value

A matrix

Examples

bcbn()

CT-CBN

Description

CT-CBN

Usage

ctcbn(
  datasets,
  bootstrapSamples = 0,
  randomSeed = 1,
  samplingRate = 1,
  epsilon = 2,
  numDrawnSamples = 0,
  numEmRuns = 1,
  nCores = 1,
  progressBar = FALSE
)

Arguments

datasets

Vector of Spock objects with poset and pattern/lambda data or a Spock object (alias of ctcbnSingle).
bootstrapSamples

Number of bootstrap samples (requires epsilon > 0, numDrawnSamples = 0)
randomSeed

Random seed.
samplingRate

Sampling rate.
epsilon

If between 0 and 1, the fraction of violations allowed per edge. If negative, the interval 0 to 0.5 will be sampled equidistantly with N points and the output Spock will include multiple resulting posets.
numDrawnSamples

If > 0, the number of samples to draw from the model. If zero (default), the model will be learned from data.
numEmRuns

Number of em runs.
nCores

Maximum number of threads to use to parallelize.
progressBar

Print out progress bar; default is FALSE

Value

A matrix of results.

Examples

examplePath <- getExamples()[3]
bc <- Spock$new(
    poset = readPoset(examplePath)$sets,
    numMutations = readPoset(examplePath)$mutations,
    genotypeMatrix = readPattern(examplePath)
)
ctcbn(bc)
ctcbn(c(bc, bc, bc))

CT-CBN Single Batch

Description

CT-CBN Single Batch

Usage

ctcbnSingle(
  dataset,
  bootstrapSamples = 0,
  randomSeed = 1,
  samplingRate = 1,
  epsilon = 2,
  numDrawnSamples = 0,
  numEmRuns = 1
)

Arguments

dataset

Spock object with poset and pattern/lambda data.
bootstrapSamples

Number of bootstrap samples (requires epsilon > 0, numDrawnSamples = 0)
randomSeed

Random seed.
samplingRate

Sampling rate.
epsilon

If between 0 and 1, the fraction of violations allowed per edge. If negative, the interval 0 to 0.5 will be sampled equidistantly with N points and the output Spock will include multiple resulting posets.
numDrawnSamples

If > 0, the number of samples to draw from the model. If zero (default), the model will be learned from data.
numEmRuns

Number of em runs.

Value

A list of output data.

Examples

examplePath <- getExamples()[1]
bc <- Spock$new(
    poset = readPoset(examplePath)$sets,
    numMutations = readPoset(examplePath)$mutations,
    genotypeMatrix = readPattern(examplePath)
)
ctcbnSingle(bc)

edgeMarginalized

Description

edgeMarginalized

Usage

edgeMarginalized(pathProb, x)

Arguments

pathProb

The pathway probabilities returned in the step 3 of the R-CBN algorithm
x

The number of mutations to consider

Value

returns the marginal probability of all the potential edges

Examples

dag <- matrix(c(2, 2, 4, 1, 3, 3), 3, 2)
lambda <- c(1, 4, 3, 2.5, 2)
x <- 4
pathP <- pathProbCBN(dag, lambda, x)
edgeMarginalized(pathP, x)

Generate Data

Description

Generate Data

Usage

generateData(poset, theta, eps, n)

Arguments

poset

Poset matrix
theta

Vector of theta values
eps

Epsilon
n

N

Value

A matrix

Examples

poset <- matrix(0, 10, 10)
poset[1, 2] <- 1
poset[2, 3] <- 1
poset[3, 4] <- 1
poset[5, 4] <- 1
poset[6, 7] <- 1
poset[8, 9] <- 1
poset[8, 10] <- 1
poset[6, 9] <- 1
tr <- transitiveClosure(poset)
theta <- c(0.8, 0.7, 0.6, 0.7, 0.4, 0.25, 0.6, 0.75, 0.5, 0.2)
eps <- 0.1
n <- 400
generateData(tr, theta, eps, n)

generateMatrixGenotypes

Description

generateMatrixGenotypes

Usage

generateMatrixGenotypes(g)

Arguments

g

genotype length

Value

a genotype matrix with ncol=g and nrow=2^g

Examples

generateMatrixGenotypes(4)

Generate TCGA Genotype Matrix

Description

Generate TCGA Genotype Matrix

Usage

generateTCGAMatrix(
  rawData = suppressMessages(getRawTCGAData("TCGA-BLCA")),
  genes = c("TP53", "ARID1A", "KDM6A", "PIK3CA", "RB1", "EP300", "FGFR3", "CREBBP",
    "STAG2", "ATM")
)

Arguments

rawData

Raw TCGA data generated using getRawTCGAData
genes

Genes to generate genotype matrix on

Value

A genotype matrix where each row is a patient and each column is a gene

Examples

generateTCGAMatrix(rawData = data.frame())
# generateTCGAMatrix()

genotypeFeasibility

Description

genotypeFeasibility

Usage

genotypeFeasibility(genotypes, dag, x)

Arguments

genotypes

the full set of potential binary genotypes of a given length.
dag

matrix representing the DAG of restrictions.
x

the number of mutations considered.

Value

a binary vector, which indicates feasibility or infeasibility of a set of genotypes

Examples

geno4 <- generateMatrixGenotypes(4)
dag <- matrix(c(4, 4, 4, 1, 2, 3), 3, 2)
x <- 4
genoF4 <- genotypeFeasibility(geno4, dag, x)

genotypeMatrixMutator

Description

genotypeMatrixMutator

Usage

genotypeMatrixMutator(mat, fp, fn)

Arguments

mat

The genotype matrix including sampled genotypes, which need to be mutated.
fp

False positive rate
fn

False negative rate

Value

The mutated version of the genotype matrix

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 800, replace = TRUE), 200, 4)
gMatMut <- genotypeMatrixMutator(gMat, 0.2, 0.2)

Get paths to examples

Description

Get paths to examples

Usage

getExamples()

Value

A vector of paths

Examples

getExamples()

Get Raw TCGA Data

Description

Get Raw TCGA Data

Usage

getRawTCGAData(project)

Arguments

project

TCGA project ID; pass "help" to see list of all project IDs

Value

data frame of TCGA data for given project

Examples

getRawTCGAData("help")

H-CBN

Description

H-CBN

Usage

hcbn(
  datasets,
  anneal = FALSE,
  temp = 0,
  annealingSteps = 0,
  epsilon = 2,
  nCores = 1,
  progressBar = FALSE
)

Arguments

datasets

Vector of Spock objects with poset and pattern/lambda data or a Spock object (alias of hcbnSingle).
anneal

If TRUE, performes a simulated annealing run starting from the poset
temp

Temperature of simulated annealing.
annealingSteps

Number of simulated annealing steps.
epsilon

Value of eps for CT-CBN model selection. Requires both pattern and lambda data in input Spock.
nCores

Maximum number of threads to use to parallelize.
progressBar

Print out progress bar; default is FALSE

Value

A matrix of results.

Examples

examplePath <- getExamples()[3]
bc <- Spock$new(
    poset = readPoset(examplePath)$sets,
    numMutations = readPoset(examplePath)$mutations,
    genotypeMatrix = readPattern(examplePath)
)
hcbn(bc)
hcbn(c(bc, bc, bc))

H-CBN Single Batch

Description

H-CBN Single Batch

Usage

hcbnSingle(
  datasetObj,
  anneal = FALSE,
  temp = 0,
  annealingSteps = 0,
  epsilon = 2
)

Arguments

datasetObj

Spock object with poset and pattern/lambda data.
anneal

If TRUE, performes a simulated annealing run starting from the poset
temp

Temperature of simulated annealing.
annealingSteps

Number of simulated annealing steps.
epsilon

Value of eps for CT-CBN model selection. Requires both pattern and lambda data in input Spock.

Value

A list of output data.

Examples

examplePath <- getExamples()[1]
bc <- Spock$new(
    poset = readPoset(examplePath)$sets,
    numMutations = readPoset(examplePath)$mutations,
    genotypeMatrix = readPattern(examplePath)
)
hcbnSingle(bc)

jensenShannonDivergence

Description

jensenShannonDivergence

Usage

jensenShannonDivergence(prob1, prob2)

Arguments

prob1

The first (discrete) probability distribution (vector)
prob2

The second (discrete) probability distribution (vector)

Value

Jensen Shannon Divergence between the two (discrete) probability distributions

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathCT <- pathProbQuartetCTCBN(gMat)
pathH <- pathProbQuartetHCBN(gMat)
jensenShannonDivergence(pathCT, pathH)

pathEdgeMapper

Description

pathEdgeMapper

Usage

pathEdgeMapper(x)

Arguments

x

number of mutations to consider

Value

Pathway to edge compatibility matrix, each element of which indicates whether a given edge is included in the transitive closure of a given pathway (1) or not (0).

Examples

peMap <- pathEdgeMapper(4)

pathNormalization

Description

pathNormalization

Usage

pathNormalization(pathProb, x)

Arguments

pathProb

The pathway probabilities returned in the step 3 of the R-CBN algorithm
x

The number of mutations to consider

Value

The updated pathway probabilities (the step 5 of the R-CBN algorithm)

Examples

dag <- matrix(c(2, 2, 4, 1, 3, 3), 3, 2)
lambda <- c(1, 4, 3, 2.5, 2)
x <- 4
pathP <- pathProbCBN(dag, lambda, x)
pathN <- pathNormalization(pathP, x)

pathProbCBN: quantifies pathway probabilities using the output of CT-CBN or H-CBN

Description

pathProbCBN: quantifies pathway probabilities using the output of CT-CBN or H-CBN

Usage

pathProbCBN(dag, lambda, x)

Arguments

dag

matrix representing the DAG of restrictions.
lambda

the lambda values, which are produced by the CBN model.
x

the number of mutations considered.

Value

vector of probabilities assigned to all potential pathways of length x

Examples

dag <- matrix(c(2, 2, 4, 1, 3, 3), 3, 2)
lambda <- c(1, 4, 3, 2.5, 2)
x <- 4
pathP <- pathProbCBN(dag, lambda, x)

pathProbQuartetBCBN

Description

pathProbQuartetBCBN

Usage

pathProbQuartetBCBN(gMat)

Arguments

gMat

The n by 4 binary genotype matrix representing a given quartet for a sample of n genotypes.

Value

The probability distribution (returned by the B-CBN model), which is represented as a vector of length 24.

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathProbQuartetBCBN(gMat)

pathProbQuartetCTCBN

Description

pathProbQuartetCTCBN

Usage

pathProbQuartetCTCBN(gMat)

Arguments

gMat

The n by 4 binary genotype matrix representing a given quartet for a sample of n genotypes.

Value

The probability distribution (returned by the CT-CBN model), which is represented as a vector of length 24.

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathProbQuartetCTCBN(gMat)

pathProbQuartetHCBN

Description

pathProbQuartetHCBN

Usage

pathProbQuartetHCBN(gMat)

Arguments

gMat

The n by 4 binary genotype matrix representing a given quartet for a sample of n genotypes.

Value

The probability distribution (returned by the H-CBN model), which is represented as a vector of length 24.

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathProbQuartetHCBN(gMat)

pathProbQuartetRCBN

Description

pathProbQuartetRCBN

Usage

pathProbQuartetRCBN(gMat)

Arguments

gMat

The n by 4 binary genotype matrix representing a given quartet for a sample of n genotypes.

Value

The probability distribution (returned by the R-CBN model), which is represented as a vector of length 24

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathProbQuartetRCBN(gMat)

pathProbSSWM

Description

pathProbSSWM

Usage

pathProbSSWM(fitness, x)

Arguments

fitness

A vector of length 2^x, each element of which representing the fitness assigned to one of the 2^x genotypes.
x

The number of mutations considered.

Value

vector of probabilities assigned to all potential pathways of length x

Examples

f <- c(0, 0.1, 0.2, 0.1, 0.2, 0.4, 0.3, 0.2, 0.2, 0.1, 0, 0.6, 0.4, 0.3, 0.2, 1)
x <- 4
pathP <- pathProbSSWM(f, x)

pathwayCompatibilityQuartet

Description

pathwayCompatibilityQuartet

Usage

pathwayCompatibilityQuartet(gMat)

Arguments

gMat

The n by 4 binary genotype matrix representing a given quartet for a sample of n genotypes.

Value

The compatibility score, which is represented as a vector of length 24, each element of which corresponds to one of the 24 pathways of length 4.

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 800, replace = TRUE), 200, 4)
pathwayCompatibilityQuartet(gMat)

pathwayFeasibility

Description

pathwayFeasibility

Usage

pathwayFeasibility(dag, x)

Arguments

dag

matrix representing the DAG of restrictions.
x

the number of mutations considered.

Value

a binary vector, which indicates feasibility or infeasibility of a set of pathways

Examples

dag <- matrix(c(4, 4, 4, 1, 2, 3), 3, 2)
x <- 4
pathwayFeasibility(dag, x)

pathwayGenotypeCompatibility

Description

pathwayGenotypeCompatibility

Usage

pathwayGenotypeCompatibility(pathway, genotype)

Arguments

pathway

a vector representing the given pathway.
genotype

a binary vector representing the given genotype.

Value

returns 1 (if the given genotype is compatible with the given pathway), and 0 otherwise

Examples

geno1 <- c(1, 0, 1, 0)
geno2 <- c(1, 1, 0, 0)
path <- c(1, 2, 3, 4)
pathwayGenotypeCompatibility(path, geno1)
pathwayGenotypeCompatibility(path, geno2)

pathwayWeightingRCBN

Description

pathwayWeightingRCBN

Usage

pathwayWeightingRCBN(edgeProb, peMap)

Arguments

edgeProb

Marginal edge probabilities
peMap

Pathway-edge compatibility matrix

Value

The pathway weights (step 4 of the R-CBN algorithm)

Examples

dag <- matrix(c(2, 2, 4, 1, 3, 3), 3, 2)
lambda <- c(1, 4, 3, 2.5, 2)
x <- 4
pathP <- pathProbCBN(dag, lambda, x)
edgeProb <- edgeMarginalized(pathP, x)
peMap <- pathEdgeMapper(4)
pathwayWeightingRCBN(edgeProb, peMap)

permutations

Description

permutations

Usage

permutations(n, r, v = 1:n, set = TRUE, repeatsAllowed = FALSE)

Arguments

n

total number of elements in the set
r

subset size
v

1:n
set

Logical flag indicating whether duplicates should be removed from the source vector v. Defaults to TRUE.
repeatsAllowed

Logical flag indicating whether the constructed vectors may include duplicated values. Defaults to FALSE.

Value

a matrix with (n!/(n-r)!) rows and r columns

Examples

perm <- permutations(4, 4)

posetWeightingRCBN

Description

posetWeightingRCBN

Usage

posetWeightingRCBN(vec)

Arguments

vec

The likelihood vector corresponding to a given set of posets

Value

The poset weight vector determined using the reciprocal ranking method

Examples

set.seed(100)
logLik <- runif(219)
w1 <- posetWeightingRCBN(logLik)

predictability

Description

predictability

Usage

predictability(prob, x)

Arguments

prob

Pathway probability vector
x

The length of genotype vectors

Value

predictability

Examples

set.seed(100)
gMat <- matrix(sample(c(0, 1), 12, replace = TRUE), 3, 4)
pathCT <- pathProbQuartetCTCBN(gMat)
pathH <- pathProbQuartetHCBN(gMat)
predC <- predictability(pathCT, 4)
predictability(pathH, 4)

Read a .lambda file

Description

Read a .lambda file

Usage

readLambda(fileStem)

Arguments

fileStem

The filename of the .lambda file without the .lambda suffix.

Value

A matrix.

Examples

bcPath <- getExamples()[1]
readLambda(bcPath)

Read a .pat file

Description

Read a .pat file

Usage

readPattern(fileStem)

Arguments

fileStem

The filename of the .pat file without the .pat suffix.

Value

A matrix.

Examples

bcPath <- getExamples()[1]
readPattern(bcPath)

Read a .poset file

Description

Read a .poset file

Usage

readPoset(fileStem)

Arguments

fileStem

The filename of the .poset file without the .poset suffix.

Value

A list containing the number of mutations and a matrix.

Examples

bcPath <- getExamples()[1]
readPoset(bcPath)

Read a .time file

Description

Read a .time file

Usage

readTime(fileStem)

Arguments

fileStem

The filename of the .time file without the .time suffix.

Value

A matrix.

Examples

bcPath <- getExamples()[1]
readPattern(bcPath)

Poset and pattern/lambda data

Description

A data class containing poset and pattern/lambda matrices.

Details

Use the read_ methods to feed data from files.

Value

a Spock object

Public fields

poset

Poset matrix.

numMutations

Number of mutations.

genotypeMatrix

Genotype matrix.

lambda

Lambda list.

Methods

Public methods

Method new()

Create a new Spock object.

Usage
Spock$new(poset, numMutations, genotypeMatrix, lambda = NULL)
Arguments
poset

Poset matrix or list of poset matrices.

numMutations

Number of mutations.

genotypeMatrix

Genotype matrix.

lambda

Lambda list.

Returns

A new Spock object.

Method getSize()

Get the number of posets.

Usage
Spock$getSize()
Returns

Number of posets.

Method getPoset()

Write poset data to a tempfile.

Usage
Spock$getPoset(index = 1)
Arguments
index

Index of poset.

Returns

File path to tempfile.

Method getSecond()

Write pattern/lambda data to a tempfile.

Usage
Spock$getSecond(n)
Arguments
n

Number of drawn samples.

Returns

File path to tempfile.

Method getPattern()

Write pattern data to a tempfile.

Usage
Spock$getPattern()
Returns

File path to tempfile.

Method getLambda()

Write lambda data to a tempfile.

Usage
Spock$getLambda()
Returns

File path to tempfile.

Method show()

Print summary information to console.

Usage
Spock$show(verbose = FALSE)
Arguments
verbose

Method prints contents as well as dimensions to console if TRUE.

Returns

Nothing.

Method clone()

The objects of this class are cloneable with this method.

Usage
Spock$clone(deep = FALSE)
Arguments
deep

Whether to make a deep clone.

Examples

examplePath <- getExamples()[1]
bc <- Spock$new(
    poset = readPoset(examplePath)$sets,
    numMutations = readPoset(examplePath)$mutations,
    genotypeMatrix = readPattern(examplePath)
)

Transitive Closure

Description

Transitive Closure

Usage

transitiveClosure(poset)

Arguments

poset

Poset matrix

Value

Poset matrix

Examples

poset <- matrix(0, 10, 10)
poset[1, 2] <- 1
poset[2, 3] <- 1
poset[3, 4] <- 1
poset[5, 4] <- 1
poset[6, 7] <- 1
poset[8, 9] <- 1
poset[8, 10] <- 1
poset[6, 9] <- 1
transitiveClosure(poset)

Visualize CBN Model

Description

Visualize CBN Model

Usage

visualizeCBNModel(
  poset,
  nodeColor = "darkgreen",
  numNodes = max(4, max(poset))
)

Arguments

poset

Poset object to visualize
nodeColor

Color of nodes in resulting graph
numNodes

Number of nodes (default is the larger number between 4 and the largest index given in the poset)

Value

Plot (gg object) visualization of CBN model

Examples

poset <- readPoset(getExamples()[1])
visualizeCBNModel(poset$sets)

Visualize Fitness Landscape

Description

Visualize Fitness Landscape

Usage

visualizeFitnessLandscape(
  fitness,
  selectNodes = NULL,
  nGenes = 4,
  lowColor = "white",
  highColor = "blue"
)

Arguments

fitness

Fitness vectors for each genotype provided in selectNodes or for all genotypes if none selected
selectNodes

Select genotypes to visualize
nGenes

Length of each genotype
lowColor

Color for wild type genotype
highColor

Color for fully mutated genotype

Value

Plot (gg object) visualization of fitness landscape

Examples

genotypes <- c(
    "0000",
    "1000",
    "0100",
    "0010",
    "0001",
    "1100",
    "1010",
    "1001",
    "0110",
    "0101",
    "0011",
    "1110",
    "1101",
    "1011",
    "0111",
    "1111"
)
#
colIntensity <- c(0, rep(0.25, 4), rep(0.5, 6), rep(0.75, 4), 1)
visualizeFitnessLandscape(colIntensity)

Visualize Pathway Probabilities

Description

Visualize Pathway Probabilities

Usage

visualizeProbabilities(
  probabilities,
  outputFile = NULL,
  geneNames = as.character(1:inverseFactorial(length(probabilities))),
  geneColors = rainbow(length(geneNames), v = 0.5),
  columnTitles = TRUE
)

Arguments

probabilities

List or matrix of probabilities for each pathway (matrix if multiple models)
outputFile

File to output to; if none provided, a plot will be returned
geneNames

Gene names; if single character, rendered in circles
geneColors

Gene colors
columnTitles

Include column titles

Value

Plot or file name

Examples

visualizeProbabilities(c(0.05, 0.03, 0.12, 0.04, 0.02, 0, 0.05, 0.04, 0.05, 0.06, 0.04, 0.02, 0.03, 0.02, 0.05, 0.03, 0.01, 0.09, 0.06, 0.04, 0, 0.08, 0.05, 0.02))

visualizeProbabilities(c(0.05, 0.03, 0.12, 0.04, 0.02, 0, 0.05, 0.04, 0.05, 0.06, 0.04, 0.02, 0.03, 0.02, 0.05, 0.03, 0.01, 0.09, 0.06, 0.04, 0, 0.08, 0.05, 0.02), geneNames = c("AAAA", "BBBB", "CCCC", "DDDD"))

mat <- matrix(c(0.1, 0.3, 0, 0.2, 0.4, 0, 0.2, 0.2, 0.1, 0, 0.2, 0.3), ncol = 2)
visualizeProbabilities(mat, columnTitles = TRUE)