METACLUSTER - an R package for context-specific expression analysis of metabolic gene clusters

Check out our hands-on tutorial on

About

METACLUSTER facilitates comprehensive condition and tissue-specific expression analysis of metabolic gene clusters based on a probabilistic framework for characterizing metabolic gene clusters using context-specific gene expression information

A) The METACLUSTER framework. B) Cluster diagram and transcriptional activity map of the arabidiol/baruol cluster (Yu et al. 2016) (C463 based on the prediction in Schlapfer et al. 2017). Colors indicate the inferred p-value of the cluster to be transcriptionally active per condition and tissue. Gray tiles indicate condition-tissue combinations that are missing in the differential expression dataset. C) Transcriptional activity map of the 317 inferred context-specific gene clusters. Color values denote the number of the transcriptionally active gene clusters per condition-tissue. Black tiles indicate condition-tissue combinations with no inferred transcriptionally active clusters.

Contact for questions.

Installation

METACLUSTER is based on R version 3.6.1. The easiest way to install METACLUSTER is through devtools (see OS specific notes on installing devtools at the end)

library(devtools)

install_github("https://github.com/mbanf/METACLUSTER", build_vignettes=TRUE,
 repos=c("http://cran.rstudio.org", "http://bioconductor.org/packages/release/bioc"),
 dependencies=TRUE)

Usage

To run the METACLUSTER with the Schlapfer et al. 2017 A.thaliana gene cluster predictions data you can download all neccessary datasets from onedrive: datasets_athaliana. If you are using personal datasets, see the required data format for "custom" datasets in section Notes.

library(METACLUSTER) # load package

setwd("/User/home/METACLUSTER_athaliana_datasets") # set working directory to the dataset files

Load individual datasets based on their filenames:

• input_format "custom", "PCF2017_enzymes_only" or "PCF2017" (default = "PCF2017_enzymes_only")
• geneCluster the gene clusters dataset
• genes a list of genes (corresponds to the rows of the differential expression datasets)
• sample_ids_differentialExpression a list of unique identifiers referencing individual condition-tissue specific differential expression experiments
• foldChange_differentialExpression differential expression data (fold changes) as a genes x differential expression experiments
• pvalue_differentialExpression differential expression data (p-values) as a genes x differential expression experiments
• experiment_condition_tissue_annotation experiment to treatment and tissue annotation (with corresponding experiment_ids)

l.data = load_datasets(input_format = "PCF2017_enzymes_only",
                      filename.geneCluster = "data/ath_geneInCluster_3_aracyc.txt-labeled_NoHypoGenes.txt",
                      filename.genes = "data/genes.txt",
                      filename.sample_ids_differentialExpression = "data/sample_ids_differentialExpression.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")

METACLUSTER Parameter sets:

!We set b.load_codifferentialAnalysis_monteCarloSimulation = "yes" for the Schlapfer et al. 2017 A.thaliana gene cluster predictions data, as we have pre-computed and provided all co-differential expression datasets - for other datasets, set to "no"!

• 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
• 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 = 100)
• min_number_condition_samples minimum number of condition samples for significance test (default 1)
• foldername.tmp temp file folder name (default = /tmp)
• foldername.results results file folder name (default = /results)

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/")

Next evaluate and store the results

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/")

Notes

Installation of devtools dependencies under Ubuntu (prior to installing devtools): sudo apt-get install build-essential libcurl4-gnutls-dev libxml2-dev libssl-dev

Then install.packages("devtools")

Custom gene cluster data format: "Cluster.ID", "Gene.ID", "Gene.Name", see custom_example_data

as a pre-requisite, our algorithm needs two matrices: m.pvalue_differentialExpression and m.foldChange_differentialExpression.

• genes a list of genes (corresponds to the rows of the differential expression datasets)
• sample_ids_differentialExpression a list of unique identifiers referencing individual condition-tissue specific differential expression experiments listed in experiment_condition_tissue_annotation (corrresponding to the columns of the differential expression datasets)

The format of the experimental annotation should be: "series_id" "condition_treatment_1" "condition_treatment_2" "condition_tissue" "unique_ID".

References

Banf M, Zhao K.M., and Rhee S. METACLUSTER - an R package for context-specific expression analysis of metabolic gene clusters, Bioinformatics, 2019

Genome-wide prediction of metabolic enzymes, pathways, and gene clusters in plants, Schläpfer P, Zhang P, Wang C, Kim T, Banf M, Chae L, Dreher K, Chavali A K, Nilo-Poyanco, Bernhard T, Kahn D, and Rhee S. - Plant physiology, 2017