run_METACLUSTER: run Metacluster algorithm

In mbanf/METACLUSTER: A package for context-specific expression analysis of metabolic gene clusters

Description

This function runs the condition specific gene cluster annotation

Usage

run_METACLUSTER(m.foldChange_differentialExpression,
  m.pvalue_differentialExpression, df.experiment_condition_annotation,
  df.geneCluster, tb.condition_treatments, tb.condition_tissues,
  n.cpus = 1,
  b.load_codifferentialAnalysis_monteCarloSimulation = "yes",
  pvalue_DifferentialExpression = 0.05,
  probability_codifferentialExpression_MonteCarloSimulation = 0.95,
  pvalue_coexpression_distribution = 0.05,
  pvalue_geneClusterPrediction = 0.05,
  pvalue_geneClusterConsistency = 0.05,
  pvalue_treatment_per_condition = 0.05,
  pvalue_tissue_per_condition = 0.05,
  number_codifferentialExpression_MonteCarloSimulations = 3,
  number_conditionSpecificCoexpressionBackgroundGenePairs = 100,
  min_number_condition_samples = 1, seed = 1234, heatmap_width = 10,
  heatmap_height = 6, foldername.tmp = "tmp/",
  foldername.results = "results/")

Arguments

m.foldChange_differentialExpression	differential expression foldchange matrix - rows are genes, cols are experiments
m.pvalue_differentialExpression	differential expression pvalue matrix - rows are genes, cols are experiments
df.experiment_condition_annotation	experiment condition annotation
df.geneCluster	gene cluster dataset
tb.condition_treatments	table of conditions
tb.condition_tissues	table of tissues
n.cpus	number of cores used (default = 1)
b.load_codifferentialAnalysis_monteCarloSimulation	load codifferential expression data ("yes", "no")
pvalue_DifferentialExpression	pvalue treshold for differential expession (default = 0.05)
probability_codifferentialExpression_MonteCarloSimulation	probability threshold codifferential expression (default = 0.05)
pvalue_coexpression_distribution	pvalue treshold context specific coexpression (default = 0.05)
pvalue_geneClusterPrediction	pvalue gene cluster inference enzyme presence (default = 0.05)
pvalue_geneClusterConsistency	pvalue gene cluster enzyme condition consistency (default = 0.05)
pvalue_treatment_per_condition	pvalue gene pair condition annotation (default = 0.05)
pvalue_tissue_per_condition	pvalue gene pair tissue annotation (default = 0.05)
number_codifferentialExpression_MonteCarloSimulations	number of codiffernetial expression background monte carlo simulations (default = 1)
number_conditionSpecificCoexpressionBackgroundGenePairs	number of context specific coexpression simulation background gene pairs (default = 50)
min_number_condition_samples	minimum number of condition samples for significance test (default 1)
heatmap_width	default = 10
heatmap_height	default = 5
foldername.tmp	temp file folder name (default = /tmp)
foldername.results	results file folder name (default = /results)
v.conditionGroups	treatment and condition map
v.tissueGroups	tissue maps
th.consistent_condition_presence_percentage	percentage of gene cluster enyzmes that are expressed in each condition in order to annotate the condition to the cluster (default = 0.7

Value

a list of results

Examples

# install_and_load_libraries()

# set directory to dataset directory, e.g. /User/home/athaliana_schlapfer2017/
setwd(...)


message("load datasets")
l.data = load_datasets(input_format = "PCF2017_enzymes_only",
                       filename.genes = "data/genes.txt",
                       filename.experiment_ids = "data/experiment_ids.txt",
                       filename.geneCluster = "data/ath_geneInCluster_3_aracyc.txt-labeled_NoHypoGenes.txt",
                       filename.foldChange_differentialExpression = "data/m.foldChange_differentialExpression.txt",
                       filename.pvalue_differentialExpression =	"data/m.pvalue_differentialExpression.txt",
                       filename.experiment_condition_tissue_annotation ="data/experiment_annotation.txt")

message("run METACLUSTER")
df.cluster_annotations = run_METACLUSTER(m.foldChange_differentialExpression = l.data$m.foldChange_differentialExpression,
                                         m.pvalue_differentialExpression = l.data$m.pvalue_differentialExpression,
                                         df.experiment_condition_annotation = l.data$df.experiment_condition_annotation,
                                         df.geneCluster = l.data$df.geneCluster,
                                         tb.condition_treatments = l.data$tb.condition_treatments,
                                         tb.condition_tissues = l.data$tb.condition_tissues,
                                         n.cpus = 3,
                                         b.load_codifferentialAnalysis_monteCarloSimulation = "yes",
                                         pvalue_DifferentialExpression = 0.05,
                                         probability_codifferentialExpression_MonteCarloSimulation = 0.95,
                                         pvalue_coexpression_distribution = 0.05,
                                         pvalue_geneClusterPrediction = 0.05,
                                         pvalue_geneClusterConsistency = 0.05,
                                         pvalue_treatment_per_condition = 0.05,
                                         pvalue_tissue_per_condition = 0.05,
                                         number_codifferentialExpression_MonteCarloSimulations = 1,
                                         number_conditionSpecificCoexpressionBackgroundGenePairs = 100,
                                         min_number_condition_samples = 1,
                                         seed = 1234,
                                         heatmap_width = 10,
                                         heatmap_height = 5,
                                         foldername.results = "results/",
                                         foldername.tmp = "tmp/")


evaluate_and_store_results(df.cluster_annotations=df.cluster_annotations,
                           df.experiment_condition_annotation = l.data$df.experiment_condition_annotation,
                           tb.condition_treatments = l.data$tb.condition_treatments,
                           tb.condition_tissues = l.data$tb.condition_tissues,
                           min_number_of_genes = 3,
                           heatmap_width = 4, heatmap_height = 7, fontsize = 7, fontsize_row = 10, fontsize_col = 10,
                           foldername.results = "results/")

mbanf/METACLUSTER documentation built on Feb. 27, 2020, 1:32 p.m.