aracne {ARACNE} R Documentation

Infer the ARACNE gene regulatory network from the expression matrix.

Description

Usage

aracne(x,
    TF = NULL,
    estimator = "copula",
    threshold = "permutation",
    mi.cutoff = 0,
    quantile = 0.95,
    permutations = 0,
    dpi = "auto",
    epsilon = 0.1,
    dpi.passes = 5,
    direction = "tf",
    bins = 4,
    max.neighbors = 0,
    max.genes = 0,
    min.nonzero = 0,
    log2 = FALSE,
    rank = FALSE,
    standardize = FALSE,
    prior = NULL,
    prior.min.confidence = 0.45,
    prior.factor = 0.5,
    prior.protection = "graded",
    prior.strength = 0.5,
    prior.tf.augment = TRUE,
    threads = -1,
    seed = 1234567,
    verbose = TRUE);

Arguments

x

the gene expression matrix, rows are the genes and columns are the samples. [as Matrix]

TF

an optional gene id vector of the transcription factors. When it is given the computation space is reduced from N x N to TF x N, and the edges are oriented from the TF to the targets; the prior network (the prior argument) is additionally used for the threshold relaxation and the DPI protection.

estimator

see the aracne_options api; default copula. [as string]

threshold

see the aracne_options api; default permutation. [as string]

mi.cutoff

the fixed MI threshold, a positive value forces the fixed mode. [as double]

quantile

the quantile of the permutation null distribution, default 0.95. [as double]

permutations

the number of the gene pairs that are sampled for the null distribution. [as integer]

dpi

auto (default), standard, tf or off. [as string]

epsilon

the DPI tolerance, default 0.1; raise it (for example 0.2 ~ 0.35) to keep more near-tie edges. [as double]

dpi.passes

the max number of the DPI passes, default 5; fewer passes keep more edges. [as integer]

direction

tf (default) or undirected. [as string]

bins

the bin number of the histogram estimator, default 4. [as integer]

max.neighbors

the top-K pruning of each regulator, 0 = no pruning. [as integer]

max.genes

keep only the top variant genes (the TF genes are always kept). [as integer]

min.nonzero

the minimal non-zero ratio of the retained genes. [as double]

log2

apply log2(x + 1) before the inference. [as boolean]

rank

spearman (rank) based copula instead of the pearson one. [as boolean]

standardize

z-score the expression vectors before the inference. [as boolean]

prior

the curated TF -> target evidence: the value of the bnlearn::prior_network api (a RegulatoryEdge vector or pipeline) can be passed directly. Only the edges whose confidence reaches prior_min_confidence are used, and the RegulationType (activator/inhibitor) is exported as the regulation column of the edge table.

prior.min.confidence

the eligibility threshold of the prior edges, default 0.45. [as double]

prior.factor

the threshold relaxation factor at the full confidence, default 0.5. [as double]

prior.protection

graded (default), sampled or off. [as string]

prior.strength

the strength of the graded protection, default 0.5. [as double]

prior.tf.augment

append the prior regulators into the TF set. [as boolean]

threads

the max degree of parallelism, a value that is less than or equals to 0 means "auto". [as integer]

seed

the random seed, the result is reproducible for a fixed seed. [as integer]

verbose

print the statistics, the top regulons and the top edges. [as boolean]

env

the R# runtime environment object. [as Environment]

Details

Authors

TRNtoolkit

Value

the fitted ARACNETools.aracneNetwork() object. Use the edges/regulons/ summary/as_graph api to inspect it, and the write_edges api to export it.

clr value class

Examples

 let net = ARACNE::aracne(hsa, TF = TF$Symbol, prior = prior, prior_protection = "graded");
 let tab = ARACNE::edges(net);
 write.csv(ARACNE::edges(net), file = "./aracne-edges.csv");

[Package ARACNE version 1.0.0.0 Index]