Package 'MOGAMUN'

Title: MOGAMUN: A Multi-Objective Genetic Algorithm to Find Active Modules in Multiplex Biological Networks
Description: MOGAMUN is a multi-objective genetic algorithm that identifies active modules in a multiplex biological network. This allows analyzing different biological networks at the same time. MOGAMUN is based on NSGA-II (Non-Dominated Sorting Genetic Algorithm, version II), which we adapted to work on networks.
Authors: Elva-María Novoa-del-Toro [aut, cre]
Maintainer: Elva-María Novoa-del-Toro <[email protected]>
License: GPL-3 + file LICENSE
Version: 1.15.0
Built: 2024-06-30 03:56:50 UTC
Source: https://github.com/bioc/MOGAMUN

Help Index

mogamun_init

Description

initialize evolution parameters

Usage

mogamun_init(
    Generations = 500,
    PopSize = 100,
    MinSize = 15,
    MaxSize = 50,
    CrossoverRate = 0.8,
    MutationRate = 0.1,
    JaccardSimilarityThreshold = 30,
    TournamentSize = 2,
    Measure = "FDR",
    ThresholdDEG = 0.05,
    MaxNumberOfAttempts = 3
)

Arguments

Generations

number of generations to run (default = 500)
PopSize

number of subnetworks in the population (default = 100)
MinSize

minimum size (no. of nodes) of the subnetworks (default = 15)
MaxSize

maximum size (no. of nodes) of the subnetworks (default = 50)
CrossoverRate

rate for the crossover (default = 0.8)
MutationRate

rate for the mutation (default = 0.1)
JaccardSimilarityThreshold

subnetworks over this Jaccard similarity threshold are considered as duplicated (default = 30)
TournamentSize

size of the tournament (default = 2)
Measure

measure to calculate the nodes scores and to determine which genes are differentially expressed (possible values PValue and FDR, default = FDR)
ThresholdDEG

threshold to consider a gene as significantly differerentially expressed. Note: if there is a logFC available, it is also considered |logFC|>1 (default = 0.05)
MaxNumberOfAttempts

maximum number of attempts to find compatible parents (default = 3)

Value

EvolutionParameters

Examples

EvolutionParameters <- 
    mogamun_init(
        Generations = 1,
        PopSize = 10,
        MinSize = 15,
        MaxSize = 50,
        CrossoverRate = 0.8,
        MutationRate = 0.1,
        JaccardSimilarityThreshold = 30,
        TournamentSize = 2,
        Measure = "FDR",
        ThresholdDEG = 0.05,
        MaxNumberOfAttempts = 3
    )

mogamun_load_data

Description

Load the data to process

Usage

mogamun_load_data(
    EvolutionParameters,
    DifferentialExpressionPath,
    NodesScoresPath,
    NetworkLayersDir,
    Layers
)

Arguments

EvolutionParameters

evolution paramenters returned by mogamun_init()
DifferentialExpressionPath

full path to the differential expression results file (in CSV format). This file must contain at least the columns "gene" with the gene names, and ("PValue" or "FDR"). It can also contain "logFC"
NodesScoresPath

full path to an existing CSV file containing the nodes scores (columns "gene" and "nodescore"). NOTE. If the file does not exist, MOGAMUN will generate it in the provided path with the specified name
NetworkLayersDir

path of the folder that contains the networks that will be the layers of the multiplex. NOTE. Each file must start with a different digit
Layers

string of numbers, where the numbers correspond to the first character of the name of the network files (e.g. "123" builds a multiplex with layers 1, 2, and 3)

Value

List with the data to process

Examples

DEGPath <- system.file("extdata/DE/Sample_DE.csv", package = "MOGAMUN")
NodesScoresPath <- 
    system.file("extdata/DE/Sample_NodesScore.csv", package = "MOGAMUN")
LayersPath <- 
    paste0(system.file("extdata/LayersMultiplex", package = "MOGAMUN"), "/")
EvolutionParameters <- mogamun_init(Generations = 1, PopSize = 10)
LoadedData <- 
    mogamun_load_data(
        EvolutionParameters = EvolutionParameters,
        DifferentialExpressionPath = DEGPath,
        NodesScoresPath = NodesScoresPath,
        NetworkLayersDir = LayersPath,
        Layers = "23"
    )

mogamun_postprocess

Description

Postprocess the results: i) calculates the accumulated Pareto front, i.e. the individuals on the first Pareto front after re-ranking the results from multiple runs (NOTE. If there is a single run, the result is the set of individuals in the first Pareto front), ii) filters the networks to leave only the interactions between the genes that are included in the results, iii) generates some plots of interest, such as scatter plots and boxplots, and iv) (optional) creates a Cytoscape file to visualize the results, merging the subnetworks with a Jaccard similarity coefficient superior to JaccardSimilarityThreshold (NOTE. Make sure to open Cytoscape if VisualizeInCytoscape is TRUE)

Usage

mogamun_postprocess(
    ExperimentDir = ".",
    LoadedData = LoadedData,
    JaccardSimilarityThreshold = 70,
    VisualizeInCytoscape = TRUE
)

Arguments

ExperimentDir

folder containing the results to be processed. It is the same folder specified as ResultsDir in mogamun_run
LoadedData

list returned by mogamun_load_data()
JaccardSimilarityThreshold

subnetworks over this Jaccard similarity threshold are merged in a single subnetwork
VisualizeInCytoscape

TRUE if you wish to visualize the accumulated Pareto front in Cytoscape, FALSE otherwise

Value

None

Examples

DEGPath <- system.file("extdata/DE/Sample_DE.csv", package = "MOGAMUN")
NodesScoresPath <- 
    system.file("extdata/DE/Sample_NodesScore.csv", package = "MOGAMUN")
LayersPath <- 
    paste0(system.file("extdata/LayersMultiplex", package = "MOGAMUN"), "/")
EvolutionParameters <- mogamun_init(Generations = 1, PopSize = 10)
LoadedData <- 
    mogamun_load_data(
        EvolutionParameters = EvolutionParameters,
        DifferentialExpressionPath = DEGPath,
        NodesScoresPath = NodesScoresPath,
        NetworkLayersDir = LayersPath,
        Layers = "23"
    )
ResultsDir <- system.file("SampleResults", package="MOGAMUN")
mogamun_run(
    LoadedData = LoadedData,
    Cores = 1,
    NumberOfRunsToExecute = 1,
    ResultsDir = ResultsDir
)
mogamun_postprocess(
    ExperimentDir = ResultsDir,
    LoadedData = LoadedData,
    JaccardSimilarityThreshold = 70,
    VisualizeInCytoscape = FALSE
)

mogamun_run

Description

Run the algorithm with the specified values for the evolution parameters

Usage

mogamun_run(LoadedData, Cores = 1, NumberOfRunsToExecute = 1, ResultsDir = ".")

Arguments

LoadedData

list returned by mogamun_load_data()
Cores

to run MOGAMUN in parallel on the given number of cores (in line with the number of physical processor cores) (default = 1)
NumberOfRunsToExecute

number of runs (default = 1)
ResultsDir

outputs the results in the specified folder

Value

None

Examples

DEGPath <- system.file("extdata/DE/Sample_DE.csv", package = "MOGAMUN")
NodesScoresPath <- 
    system.file("extdata/DE/Sample_NodesScore.csv", package = "MOGAMUN")
LayersPath <- 
    paste0(system.file("extdata/LayersMultiplex", package = "MOGAMUN"), "/")
EvolutionParameters <- mogamun_init(Generations = 1, PopSize = 10)
LoadedData <- 
    mogamun_load_data(
        EvolutionParameters = EvolutionParameters,
        DifferentialExpressionPath = DEGPath,
        NodesScoresPath = NodesScoresPath,
        NetworkLayersDir = LayersPath,
        Layers = "23"
    )
ResultsDir <- system.file("SampleResults", package="MOGAMUN")
mogamun_run(
    LoadedData = LoadedData,
    Cores = 1,
    NumberOfRunsToExecute = 1,
    ResultsDir = ResultsDir
)