Package 'GeDi'

Title: Defining and visualizing the distances between different genesets
Description: The package provides different distances measurements to calculate the difference between genesets. Based on these scores the genesets are clustered and visualized as graph. This is all presented in an interactive Shiny application for easy usage.
Authors: Annekathrin Nedwed [aut, cre] , Federico Marini [aut]
Maintainer: Annekathrin Nedwed <[email protected]>
License: MIT + file LICENSE
Version: 1.3.0
Built: 2024-10-30 07:28:10 UTC
Source: https://github.com/bioc/GeDi

Help Index

Check genesets format

Description

Check if the input genesets have the expected format for this app

Usage

.checkGenesets(
  genesets,
  col_name_genesets = "Genesets",
  col_name_genes = "Genes"
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
col_name_genesets

character, the name of the column in which the geneset ids are listed. Defaults to "Genesets".
col_name_genes

character, the name of the column in which the genes are listed. Defaults to "Genes".

Value

A validated and formatted genesets data frame.

Check GeneTonic List format

Description

Check if the provided GeneTonic List object has the expected format for the app and extract the functional enrichment results

Usage

.checkGTL(gtl)

Arguments

gtl

A GeneTonicListobject generated with GeneTonic::GeneTonic_list(), containing the functional enrichment results.

Value

A validated and renamed geneset data.frame.

Check PPI format

Description

Check if the Protein-Protein-interaction (PPI) has the expected format for this app

Usage

.checkPPI(ppi)

Arguments

ppi

a data.frame, Protein-protein interaction (PPI) network data frame. The object is expected to have three columns, Gene1 and Gene2 which specify the gene names of the interacting proteins in no particular order (symmetric interaction) and a column combined_score which is a numerical value of the strength of the interaction.

Value

A validated and formatted PPI data frame.

Check distance scores format

Description

Check if the provided distance scores have the expected format for this app

Usage

.checkScores(genesets, distance_scores)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
distance_scores

A Matrix::Matrix() or object, A matrix with numerical (distance) scores.

Value

A validated and formatted distance_scores Matrix::Matrix().

Filter Genesets from the input data

Description

Filter a preselected list of genesets from a data.frame of genesets

Usage

.filterGenesets(remove, df_genesets)

Arguments

remove

a list, A list of geneset names to be removed
df_genesets

a data.frame, A data.frame with at least two columns. One should be called Geneset, containing the names/identifiers of the genesets in the data. The second column should be called Genes and contains one string of the genes contained in each geneset.

Value

A data.frame containing information about filtered genesets

Make an educated guess on the separator character

Description

This function tries to guess which separator was used in a list of delimited strings.

Usage

.findSeparator(stringList, sepList = c(",", "\t", ";", " ", "/"))

Arguments

stringList

list, a list of strings
sepList

list, containing the candidates for being identified as separators. Defaults to c(",", "\t", ";"," ", "/").

Value

character, corresponding to the guessed separator. One of "," (comma), "\t" (tab), ";" (semicolon)," " (whitespace) or "/" (backslash).

References

See https://github.com/federicomarini/ideal for details on the original implementation.

Map each geneset to the cluster it belongs

Description

Map each geneset to the cluster it belongs and return the information as a data.frame

Usage

.getClusterDatatable(cluster, gs_names, gs_description)

Arguments

cluster

A list of clusters
gs_names

A vector of geneset names
gs_description

A vector of descriptions for each geneset

Value

A data.frame mapping each geneset to the cluster(s) it belongs to

Get gene set descriptions

Description

Extracts gene set descriptions from a provided gene set object. The function prioritizes columns "Term", "Description", or "Genesets" to find the appropriate descriptions. If any descriptions are duplicated, the function appends a suffix to make them unique.

Usage

.getGenesetDescriptions(genesets)

Arguments

genesets

a data.frame, A data.frame with at least two columns. One should be called Geneset, containing the names/identifiers of the genesets in the data. The second column should be called Genes and contains one string of the genes contained in each geneset.

Value

a list of geneset descriptions

Determine the number of cores to use for a function

Description

Determine the number of CPU cores the scoring functions should use when computing the distance scores.

Usage

.getNumberCores(n_cores = NULL)

Arguments

n_cores

numeric, number of cores to use for the function. Defaults to Null in which case the function takes half of the available cores.

Value

Number of CPU cores to be used.

Generate a data.frame of graph metrics

Description

Generate a data.frame of the graph metrics degree, betweenness, harmonic centrality and clustering coefficient for each node in a given graph.

Usage

.graphMetricsGenesetsDT(g, genesets)

Arguments

g

A igraph graph object
genesets

A data.frame of genesets with a column Genesets containing geneset identifiers and a column Genes containing the genes belonging to each geneset

Value

A data.frame of geneset extended by columns for the degree, betweenness, harmonic centrality and clustering coefficient for each geneset.

Make an educated guess on the separator character

Description

This function tries to guess which separator was used in a text delimited file.

Usage

.sepguesser(file, sep_list = c(",", "\t", ";", " ", "/"))

Arguments

file

a character, location of a file to read data from.
sep_list

a list, containing the candidates for being identified as separators. Defaults to c(",", "\t", ";"," ", "/").

Value

A character, corresponding to the guessed separator. One of "," (comma), "\t" (tab), ";" (semicolon)," " (whitespace) or "/" (backslash).

References

See https://github.com/federicomarini/ideal for details on the original implementation.

Build a cluster graph

Description

Build a igraph from cluster information, connecting nodes which belong to the same cluster.

Usage

buildClusterGraph(
  cluster,
  geneset_df,
  gs_ids,
  color_by = NULL,
  gs_names = NULL
)

Arguments

cluster

list, a list of clusters, where each cluster member is indicated by a numeric value.
geneset_df

data.frame, a data.frame of genesets with at least two columns, one called Genesets containing geneset identifiers and one called Genes containing a list of genes belonging to the individual genesets.
gs_ids

vector, a vector of geneset identifiers, e.g. the Genesets column of geneset_df.
color_by

character, a column name of geneset_df which is used to color the nodes of the resulting graph. The column should ideally contain a numeric measurement. Defaults to NULL and nodes will remain uncolored.
gs_names

vector, a vector of geneset descriptions/names, e.g. the Term / Description column of geneset_df.

Value

An igraph object to be further manipulated or processed/plotted (e.g. via igraph::plot.igraph() or visNetwork::visIgraph())

Examples

cluster <- list(c(1:5), c(6:9, 1))
genes <- list(
  c("PDHB", "VARS2"), c("IARS2", "PDHA1"),
  c("AAAS", "ABCE1"), c("ABI1", "AAR2"), c("AATF", "AMFR"),
  c("BMS1", "DAP3"), c("AURKAIP1", "CHCHD1"), c("IARS2"),
  c("AHI1", "ALMS1")
)
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h", "i")
gs_ids <- c(1:9)
geneset_df <- data.frame(
  Genesets = gs_names,
  value = rep(1, 9)
)
geneset_df$Genes <- genes
graph <- buildClusterGraph(
  cluster = cluster,
  geneset_df = geneset_df,
  gs_ids = gs_ids,
  color_by = "value",
  gs_names = gs_names
)

Construct a graph

Description

Construct a graph from a given adjacency matrix

Usage

buildGraph(adjMatrix, geneset_df = NULL, gs_names = NULL, weighted = FALSE)

Arguments

adjMatrix

A Matrix::Matrix() indicating for which pair of nodes an edge should be added; 1 indicating an edge, 0 indicating no edge.
geneset_df

data.frame, a data.frame of genesets with at least two columns, one called Genesets containing geneset identifiers and one called Genes containing a list of genes belonging to the individual genesets.
gs_names

vector, a vector of geneset descriptions/names, e.g. the Term / Description column of geneset_df.
weighted

logical value, whether or not the resulting graph should have weighted edges. If TRUE, the adjMatrix values will be used as weights. Default to FALSE.

Value

An igraph object to be further manipulated or processed/plotted (e.g. via igraph::plot.igraph() or visNetwork::visIgraph())

Examples

adj <- Matrix::Matrix(0, 100, 100)
adj[c(80:100), c(80:100)] <- 1
geneset_names <- as.character(stats::runif(100, min = 0, max = 1))
rownames(adj) <- colnames(adj) <- geneset_names
graph <- buildGraph(adj)

Prepare data for gsHistogram().

Description

Prepare the data for the gsHistogram() by generating a data.frame which maps geneset names / identifiers to the size of their size.

Usage

buildHistogramData(
  genesets,
  gs_names,
  gs_description = NULL,
  start = 0,
  end = 0
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
gs_names

character vector, Name / identifier of the genesets in genesets
gs_description

Optional, a character vector containing a short description for each geneset
start

numeric, Optional, describes the minimum gene set size to include. Defaults to 0.
end

numeric, Optional, describes the maximum gene set size to include. Defaults to 0.

Value

A data.frame mapping geneset names to sizes

Examples

## Mock example showing how the data should look like
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h", "i")
genesets <- list(
  c("PDHB", "VARS2"), c("IARS2", "PDHA1"),
  c("AAAS", "ABCE1"), c("ABI1", "AAR2"), c("AATF", "AMFR"),
  c("BMS1", "DAP3"), c("AURKAIP1", "CHCHD1"), c("IARS2"),
  c("AHI1", "ALMS1")
)

p <- buildHistogramData(genesets, gs_names)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
p <- buildHistogramData(genes, macrophage_topGO_example_small$Genesets)

Calculate the Jaccard distance

Description

Calculate the Jaccard distance between two genesets.

Usage

calculateJaccard(a, b)

Arguments

a, b

character vector, set of gene identifiers.

Value

The Jaccard distance of the sets.

Examples

## Mock example showing how the data should look like
a <- c("PDHB", "VARS2")
b <- c("IARS2", "PDHA1")
c <- calculateJaccard(a, b)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
jaccard <- calculateJaccard(genes[1], genes[2])

Calculate the Kappa distance

Description

Calculate the Kappa distance between two genesets.

Usage

calculateKappa(a, b, all_genes)

Arguments

a, b

character vector, set of gene identifiers.
all_genes

character vector, list of all (unique) genes available in the input data.

Value

The Kappa distance of the sets.

Examples

## Mock example showing how the data should look like
a <- c("PDHB", "VARS2")
b <- c("IARS2", "PDHA1")
all_genes <- c("PDHB", "VARS2", "IARS2", "PDHA1")
c <- calculateKappa(a, b, all_genes)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
c <- calculateKappa(genes[1], genes[2], unique(genes))

Calculate the Sorensen-Dice distance

Description

Calculate the Sorensen-Dice distance between two genesets.

Usage

calculateSorensenDice(a, b)

Arguments

a, b

character vector, set of gene identifiers.

Value

The Sorensen-Dice distance of the sets.

Examples

#' ## Mock example showing how the data should look like
a <- c("PDHB", "VARS2")
b <- c("IARS2", "PDHA1")
c <- calculateSorensenDice(a, b)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
sd <- calculateSorensenDice(genes[1], genes[2])

Check for subset inclusion

Description

Remove subsets from a given list of sets, i.e. remove sets which are completely contained in any other larger set in the list.

Usage

checkInclusion(seeds)

Arguments

seeds

A list of sets

Value

A list of unique sets

Examples

## Mock example showing how the data should look like

seeds <- list(c(1:5), c(2:5), c(6:10))
s <- checkInclusion(seeds)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

seeds <- seedFinding(scores_macrophage_topGO_example_small,
                     simThreshold = 0.3,
                     memThreshold = 0.5)
seeds <- checkInclusion(seeds)

Cluster genesets.

Description

This function performs clustering on a set of scores using either the Louvain or Markov method.

Usage

clustering(scores, threshold, cluster_method = "louvain")

Arguments

scores

A Matrix::Matrix() of (distance) scores
threshold

numerical, A threshold used to determine which genesets are considered similar. Genesets are considered similar if (distance) score <= threshold. similar.
cluster_method

character, the clustering method to use. The options are louvain and markov. Defaults to louvain.

Value

A list of clusters

Examples

## Mock example showing how the data should look like
m <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(m) <- colnames(m) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
cluster <- clustering(m, 0.3, "markov")

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

clustering <- clustering(scores_macrophage_topGO_example_small,
                        threshold = 0.5)

Deprecated functions in GeDi

Description

Functions that are on their way to the function afterlife. Their successors are also listed.

Arguments

...

Ignored arguments.

Details

The successors of these functions are likely coming from a renaming of the functions to more intuitive function names

Value

All functions throw a warning, with a deprecation message pointing towards its descendent (if available).

Renaming funciton with more intuitive names

Author(s)

Annekathrin Nedwed

Examples

# try(getGenes())

Plot a dendrogram

Description

Plot a dendrogram of a matrix of (distance) scores.

Usage

distanceDendro(distance_scores, cluster_method = "average")

Arguments

distance_scores

A Matrix::Matrix() containing (distance) scores between 0 and 1.
cluster_method

character, indicating the clustering method for the stats::hclust() function. See the stats::hclust() function for the available options. Defaults to 'average'.

Value

A ggdendro::ggdendrogram() plot object.

Examples

## Mock example showing how the data should look like

distance_scores <- Matrix::Matrix(0.5, 20, 20)
distance_scores[c(11:15), c(2:6)] <- 0.2
dendro <- distanceDendro(distance_scores, cluster_method = "single")

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
dendro <- distanceDendro(scores_macrophage_topGO_example_small,
                        cluster_method = "average")

Plot a heatmap

Description

Plot a heatmap of a matrix of (distance) scores of the input genesets

Usage

distanceHeatmap(
  distance_scores,
  chars_limit = 50,
  plot_labels = TRUE,
  cluster_rows = TRUE,
  cluster_columns = TRUE,
  title = "Distance Scores"
)

Arguments

distance_scores

A Matrix::Matrix() of (distance) scores for each pairwise combination of genesets.
chars_limit

Numeric value, Indicates how many characters of the row and column names of distance_scores should be plotted. Defaults to 50 and prevents crowded axes due to long names.
plot_labels

Logical, Indicates if row and collabels should be plotted. Defaults to TRUE
cluster_rows

Logical, Indicates whether or not the rows should be clustered based on the distance scores. Defaults to TRUE
cluster_columns

Logical, Indicates whether or not the rows should be clustered based on the distance scores. Defaults to TRUE
title

character, a title for the figure. Defaults to "Distance Scores"

Value

A ComplexHeatmap::Heatmap() plot object.

Examples

## Mock example showing how the data should look like

distance_scores <- Matrix::Matrix(0.5, 20, 20)
distance_scores[c(11:15), c(2:6)] <- 0.2
rownames(distance_scores) <- colnames(distance_scores) <- as.character(c(1:20))
p <- distanceHeatmap(distance_scores)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
p <- distanceHeatmap(scores_macrophage_topGO_example_small)

Visualize the results of an enrichment analysis as word cloud

Description

Visualize the results of an enrichment analysis as a word cloud. The word cloud highlights the most frequent terms associated with the description of the genesets in the enrichment analysis.

Usage

enrichmentWordcloud(genesets_df)

Arguments

genesets_df

A data.frame object of an enrichment analysis results. This object should follow the input requirements of GeDi(), check out the vignette for further details. Besides the specified required columns, the object should ideally include a column with a short geneset description which is used for the word cloud. If no such column is available, the row names of the data.frame are used for the word cloud.

Value

A wordcloud2::wordcloud2() plot object

Examples

## Mock example showing how the data should look like

## If no "Term" or "Description" column is available,
## the rownames of the data frame will be used.
geneset_df <- data.frame(
              Genesets = c("GO:0002503", "GO:0045087", "GO:0019886"),
              Genes = c("B2M, HLA-DMA, HLA-DMB",
                        "ACOD1, ADAM8, AIM2",
                        "B2M, CD74, CTSS")
)
rownames(geneset_df) <- geneset_df$Genesets

wordcloud <- enrichmentWordcloud(geneset_df)

## With available "Term" column.
geneset_df <- data.frame(
              Genesets = c("GO:0002503", "GO:0045087", "GO:0019886"),
              Genes = c("B2M, HLA-DMA, HLA-DMB",
                        "ACOD1, ADAM8, AIM2",
                        "B2M, CD74, CTSS"),
              Term = c(
              "peptide antigen assembly with MHC class II protein complex",
              "innate immune response",
              "antigen processing and presentation of exogenous
               peptide antigen via MHC class II")
)

wordcloud <- enrichmentWordcloud(geneset_df)


## Example using the data available in the package

data(macrophage_topGO_example,
     package = "GeDi",
     envir = environment())
wordcloud <- enrichmentWordcloud(macrophage_topGO_example)

Find cluster from initial seeds

Description

Merge the initially determined seeds to clusters.

Usage

fuzzyClustering(seeds, threshold)

Arguments

seeds

A list of seeds, e.g. determined by GeDi::seedFinding() function
threshold

numerical, A threshold for merging seeds

Value

A list of clusters

References

See https://david.ncifcrf.gov/helps/functional_classification.html#clustering for details on the original implementation

Examples

## Mock example showing how the data should look like

seeds <- list(c(1:5), c(6:10))
cluster <- fuzzyClustering(seeds, 0.5)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

seeds <- seedFinding(scores_macrophage_topGO_example_small,
                     simThreshold = 0.3,
                     memThreshold = 0.5)
cluster <- fuzzyClustering(seeds, threshold = 0.5)

GeDi main function

Description

GeDi main function

Usage

GeDi(
  genesets = NULL,
  ppi_df = NULL,
  distance_scores = NULL,
  gtl = NULL,
  col_name_genesets = "Genesets",
  col_name_genes = "Genes"
)

Arguments

genesets

a data.frame, The input data used for GeDi. This should be a data.frame of at least two columns.One column should be called "Genesets" and contain some sort of identifiers for the individual genesets. In this application, we use the term "Genesets" to refer to collections of individual genes, which share common biological characteristics or functions. Such genesets can for example be obtained from databases such as the Gene Ontology (GO), the Kyoto Encyclopedia of Genes and Genomes (KEGG), Reactome, or the Molecular Signatures Database (MSigDB). The identifiers used in these databases can be directly used as geneset identifiers in GeDi. The second column should be called "Genes" and contain a list of genes belonging to the individual genesets in the "Genesets" column. In order to leverage all of the functionality available in GeDi, the column has to contain gene names and no other commonly used identifiers. The column names are case sensitive.
ppi_df

a data.frame, Protein-protein interaction (PPI) network data frame. The object is expected to have three columns, Gene1 and Gene2 which specify the gene names of the interacting proteins in no particular order (symmetric interaction) and a column combined_score which is a numerical value of the strength of the interaction.
distance_scores

A Matrix::Matrix() of (distance) scores
gtl

A GeneTonicListobject generated with GeneTonic::GeneTonic_list(), containing the functional enrichment results.
col_name_genesets

character, the name of the column in which the geneset ids are listed. Defaults to "Genesets".
col_name_genes

character, the name of the column in which the genes are listed. Defaults to "Genes".

Value

A Shiny app object is returned

Examples

if (interactive()) {
  GeDi()
}
# Alternatively, you can also start the application with your data directly
# loaded.

data("macrophage_topGO_example", package = "GeDi")
if (interactive()) {
  GeDi(genesets = macrophage_topGO_example)
}

Construct an adjacency matrix

Description

Construct an adjacency matrix from the (distance) scores and a given threshold.

Usage

getAdjacencyMatrix(distanceMatrix, cutOff, weighted = FALSE)

Arguments

distanceMatrix

A Matrix::Matrix() containing (distance) scores between 0 and 1.
cutOff

Numeric value, indicating for which pair of entries in the distanceMatrix a 1 should be inserted in the adjacency matrix. A 1 is inserted when for each entry in the matrix that is smaller or equal to the cutOff value.
weighted

logical value, indicating whether or not the resulting adjacency matrix should be weighted. If TRUE, the matrix will be weighted by the distance scores in distanceMatrix. Defaults to FALSE.

Value

A Matrix::Matrix() of adjacency status

Examples

m <- Matrix::Matrix(stats::runif(1000, 0, 1), 100, 100)
geneset_names <- as.character(stats::runif(100, min = 0, max = 1))
rownames(m) <- colnames(m) <- geneset_names
threshold <- 0.3
adj <- getAdjacencyMatrix(m, threshold)

Get the annotation of a STRINGdb object

Description

Get the annotation of a STRINGdb object, i.e. the aliases of the protein information

Usage

getAnnotation(stringdb)

Arguments

stringdb

the STRINGdb object

Value

A data.frame mapping STRINGdb ids to gene names

Examples

stringdb <- getStringDB(9606)
stringdb
anno_df <- getAnnotation(stringdb)

Construct a bipartite graph

Description

Construct a bipartite graph from cluster information, mapping the cluster to its members

Usage

getBipartiteGraph(cluster, gs_names, genes)

Arguments

cluster

list, a list of clusters, cluster members are indicated by numeric values.
gs_names

vector, a vector of (geneset) identifiers/names to map the numeric member value in cluster to.
genes

list, a list of vectors of genenames which belong to the genesets in gs_names.

Value

An igraph object to be further manipulated or processed/plotted (e.g. via igraph::plot.igraph() or visNetwork::visIgraph())

Examples

cluster <- list(c(1:5), c(6:9))
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h", "i")
genes <- list(
  c("PDHB", "VARS2"), c("IARS2", "PDHA1"),
  c("AAAS", "ABCE1"), c("ABI1", "AAR2"), c("AATF", "AMFR"),
  c("BMS1", "DAP3"), c("AURKAIP1", "CHCHD1"), c("IARS2"),
  c("AHI1", "ALMS1")
)

g <- getBipartiteGraph(cluster, gs_names, genes)

Construct an adjacency matrix

Description

Construct an adjacency matrix from a list of cluster.

Usage

getClusterAdjacencyMatrix(cluster, gs_names)

Arguments

cluster

A list of clusters, where each cluster member is indicated by a numeric value
gs_names

A vector of geneset names

Value

A Matrix::Matrix() of adjacency status

Examples

cluster <- list(c(1:5), c(6:9))
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h", "i")
adj <- getClusterAdjacencyMatrix(cluster, gs_names)

Split string of genes

Description

Split a long string of space separated genes into a list of individual genes.

Usage

getGenes(genesets, gene_name = NULL)

Arguments

genesets

a data.frame, A data.frame with at least two columns. One should be called Geneset, containing the names/identifiers of the genesets in the data. The second column should be called Genes and contains one string of the genes contained in each geneset.
gene_name

a character, Alternative name for the column containing the genes in genesets. If not given, the column is expected to be called Genes.

Value

A list containing for each geneset in the Geneset column a list of the included genes.

Build up the node title

Description

Build up the title for the graph nodes to display the available information of each geneset.

Usage

getGraphTitle(
  geneset_df = NULL,
  node_ids,
  gs_ids,
  gs_names = NULL,
  cluster_id = NULL
)

Arguments

geneset_df

A data.frame of genesets with a column Genesets containing geneset identifiers and a column Genes containing the genes belonging to each geneset
node_ids

vector, a vector of ids of the nodes in the graph for which the node title should be build.
gs_ids

vector, a vector of geneset identifiers, e.g. the Genesets column of geneset_df.
gs_names

vector, a vector of geneset descriptions/names, e.g. the Term / Description column of geneset_df.
cluster_id

vector, a vector of cluster ids for each of the genesets

Value

A list of titles for a graph with nodes given by node_ids.

Examples

genes <- list(
  c("PDHB", "VARS2"), c("IARS2", "PDHA1"),
  c("AAAS", "ABCE1"), c("ABI1", "AAR2"), c("AATF", "AMFR"),
  c("BMS1", "DAP3"), c("AURKAIP1", "CHCHD1"), c("IARS2"),
  c("AHI1", "ALMS1")
)
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h", "i")
geneset_df <- data.frame(
  Genesets = gs_names,
  value = rep(1, 9)
)
geneset_df$Genes <- genes
graph <- getGraphTitle(
  geneset_df = geneset_df,
  node_ids = c(1:9),
  gs_ids = c(1:9),
  gs_names = gs_names
)

Get NCBI ID

Description

Get the NCBI ID of a species

Usage

getId(species, version = "12.0", cache = FALSE)

Arguments

species

character, the species of your input data
version

character, the version of STRING you want to use, defaults to the current version of STRING
cache

Logical value, defining whether to use the BiocFileCache for retrieval of the files underlying the STRINGdb object. Defaults to TRUE.

Value

A character of the NCBI ID of species

Examples

species <- "Homo sapiens"
id <- getId(species = species)

species <- "Mus musculus"
id <- getId(species = species)

Calculate interaction score for two genesets

Description

The function calculates an interaction score between two sets of genes based on a protein-protein interaction network.

Usage

getInteractionScore(a, b, ppi, maxInteract)

Arguments

a, b

character vector, set of gene identifiers.
ppi

a data.frame, Protein-protein interaction (PPI) network data frame. The object is expected to have three columns, Gene1 and Gene2 which specify the gene names of the interacting proteins in no particular order (symmetric interaction) and a column combined_score which is a numerical value of the strength of the interaction.
maxInteract

numeric, Maximum interaction value in the PPI.

Value

Interaction score between the two gene sets.

References

See https://doi.org/10.1186/s12864-019-5738-6 for details on the original implementation.

Examples

## Mock example showing how the data should look like
a <- c("PDHB", "VARS2", "IARS2")
b <- c("IARS2", "PDHA1")

ppi <- data.frame(
  Gene1 = c("PDHB", "VARS2", "IARS2"),
  Gene2 = c("IARS2", "PDHA1", "CD3"),
  combined_score = c(0.5, 0.2, 0.1)
)
maxInteract <- max(ppi$combined_score)

interaction <- getInteractionScore(a, b, ppi, maxInteract)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
data(ppi_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
maxInteract <- max(ppi_macrophage_topGO_example_small$combined_score)

interaction <- getInteractionScore(genes[1], genes[2], ppi, maxInteract)

Get Matrix of Jaccard distances

Description

Calculate the Jaccard distance of all combinations of genesets in a given data set of genesets.

Usage

getJaccardMatrix(
  genesets,
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
progress

a shiny::Progress() object, Optional progress bar object to track the progress of the function (e.g. in a Shiny app).
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with Jaccard distance rounded to 2 decimal places.

Examples

## Mock example showing how the data should look like
genesets <- list(list("PDHB", "VARS2"), list("IARS2", "PDHA1"))
m <- getJaccardMatrix(genesets)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
jaccard <-getJaccardMatrix(genes)

Get Matrix of Kappa distances

Description

Calculate the Kappa distance of all combinations of genesets in a given data set of genesets. The Kappa distance is normalized to the (0, 1) interval.

Usage

getKappaMatrix(
  genesets,
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
progress

a shiny::Progress() object, Optional progress bar object to track the progress of the function (e.g. in a Shiny app).
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with Kappa distance rounded to 2 decimal places.

Examples

#' ## Mock example showing how the data should look like
genesets <- list(list("PDHB", "VARS2"), list("IARS2", "PDHA1"))
m <- getKappaMatrix(genesets)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
kappa <- getKappaMatrix(genes)

Get Matrix of Meet-Min distances

Description

Calculate the Meet-Min distance of all combinations of genesets in a given data set of genesets.

Usage

getMeetMinMatrix(
  genesets,
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
progress

a shiny::Progress() object, Optional progress bar object to track the progress of the function (e.g. in a Shiny app).
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with Meet-Min distance rounded to 2 decimal places.

Examples

## Mock example showing how the data should look like
genesets <- list(list("PDHB", "VARS2"), list("IARS2", "PDHA1"))
m <- getMeetMinMatrix(genesets)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
mm <- getMeetMinMatrix(genes)

Calculate the pMM distance

Description

Calculate the pMM distance of all combinations of genesets in a given data set of genesets.

Usage

getpMMMatrix(
  genesets,
  ppi,
  alpha = 1,
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
ppi

a data.frame, Protein-protein interaction (PPI) network data frame. The object is expected to have three columns, Gene1 and Gene2 which specify the gene names of the interacting proteins in no particular order (symmetric interaction) and a column combined_score which is a numerical value of the strength of the interaction.
alpha

numeric, Scaling factor for controlling the influence of the interaction score. Defaults to 1.
progress

a shiny::Progress() object, Optional progress bar object to track the progress of the function (e.g. in a Shiny app).
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with pMM distance rounded to 2 decimal places.

References

See https://doi.org/10.1186/s12864-019-5738-6 for details on the original implementation.

Examples

## Mock example showing how the data should look like
genesets <- list(c("PDHB", "VARS2"), c("IARS2", "PDHA1"))

ppi <- data.frame(
  Gene1 = c("PDHB", "VARS2"),
  Gene2 = c("IARS2", "PDHA1"),
  combined_score = c(0.5, 0.2)
)

pMM <- getpMMMatrix(genesets, ppi)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
data(ppi_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

pMM <- getpMMMatrix(genes, ppi)

Download Protein-Protein Interaction (PPI)

Description

Download the Protein-Protein Interaction (PPI) information of a STRINGdb object

Usage

getPPI(genes, stringdb, anno_df)

Arguments

genes

a list, A list of genes to download the respective protein- protein interaction information
stringdb

A STRINGdb object, the species of the object should match the species of genes.
anno_df

An annotation data.frame mapping STRINGdb ids to gene names, e.g. downloaded with GeDi::getAnnotation()

Value

A data.frame of Protein-Protein interactions

Examples

## Mock example showing how the data should look like

genes <- c(c("CFTR", "RALA"), c("CACNG3", "ITGA3"), c("DVL2"))
stringdb <- getStringDB(9606, cache_location = FALSE)
# stringdb
anno_df <- getAnnotation(stringdb)
ppi <- getPPI(genes, stringdb, anno_df)

## Example using the data available in the package
## Not run: 
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
stringdb <- getStringDB(9606)
stringdb
anno_df <- getAnnotation(stringdb)
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
ppi <- getPPI(genes, stringdb, anno_df)

## End(Not run)

Get Matrix of Sorensen-Dice distances

Description

Calculate the Sorensen-Dice distance of all combinations of genesets in a given data set of genesets.

Usage

getSorensenDiceMatrix(
  genesets,
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
progress

a shiny::Progress() object, Optional progress bar object to track the progress of the function (e.g. in a Shiny app).
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with Sorensen-Dice distance rounded to 2 decimal places.

Examples

## Mock example showing how the data should look like
genesets <- list(list("PDHB", "VARS2"), list("IARS2", "PDHA1"))
m <- getSorensenDiceMatrix(genesets)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
sd_matrix <- getSorensenDiceMatrix(genes)

Get the STRING db entry of a species

Description

Get the respective STRINGdb object of your species of interest

Usage

getStringDB(
  species,
  version = "12.0",
  score_threshold = 0,
  cache_location = FALSE
)

Arguments

species

numeric, the NCBI ID of the species of interest
version

character, The STRINGdb version to use, defaults to the current version
score_threshold

numeric, A score threshold to cut the retrieved interactions, defaults to 0 (all interactions)
cache_location

Logical value, defining whether to use the BiocFileCache for retrieval of the files underlying the STRINGdb object. Defaults to TRUE.

Value

a STRINGdb object of species

Examples

species <- getId(species = "Homo sapiens")
stringdb <- getStringDB(as.numeric(species))

Calculate similarity of GO terms

Description

Calculate the pairwise similarity of GO terms

Usage

goDistance(
  geneset_ids,
  method = "Wang",
  ontology = "BP",
  species = "org.Hs.eg.db",
  progress = NULL,
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

geneset_ids

list, a list of GO identifiers to score
method

character, the method to calculate the GO distance. See GOSemSim::goSim measure parameter for possibilities.
ontology

character, the ontology to use. See GOSemSim::goSim ont parameter for possibilities.
species

character, the species of your data. Indicated as org.XX.eg.db package from Bioconductor.
progress

shiny::Progress() object, optional. To track the progress of the function (e.g. in a Shiny app)
BPPARAM

A BiocParallel bpparam object specifying how parallelization should be handled. Defaults to BiocParallel::SerialParam()

Value

A Matrix::Matrix() with the pairwise GO distance of each geneset pair.

Examples

## Mock example showing how the data should look like
go_ids <- c("GO:0002503", "GO:0045087", "GO:0019886",
            "GO:0002250", "GO:0001916", "GO:0019885")

similarity <- goDistance(go_ids)

## Example using the data available in the package
data(macrophage_topGO_example_small, package = "GeDi")
go_ids <- macrophage_topGO_example_small$Genesets
## Not run: 
similarity <- goDistance(go_ids)

## End(Not run)

Create a histogram plot for gene set sizes

Description

Create a histogram plot to plot geneset names / identifiers against their size.

Usage

gsHistogram(
  genesets,
  gs_names,
  gs_description = NULL,
  start = 0,
  end = 0,
  binwidth = 5,
  color = "#0092AC"
)

Arguments

genesets

a list, A list of genesets where each genesets is represented by list of genes.
gs_names

character vector, Name / identifier of the genesets in genesets
gs_description

Optional, a character vector containing a short description for each geneset
start

numeric, Optional, describes the minimum gene set size to include. Defaults to 0.
end

numeric, Optional, describes the maximum gene set size to include. Defaults to 0.
binwidth

numeric, Width of histogram bins. Defaults to 5.
color

character, Fill color for histogram bars. Defaults to #0092AC.

Value

A ggplot2::ggplot() plot object.

Examples

## Mock example showing how the data should look like
gs_names <- c("a", "b", "c", "d", "e", "f", "g", "h")
genesets <- list(
  c("PDHB", "VARS2", "IARS2", "PDHA1"),
  c("AAAS", "ABCE1"), c("ABI1", "AAR2", "AATF"), c("AMFR"),
  c("BMS1", "DAP3"), c("AURKAIP1", "CHCHD1"), c("IARS2"),
  c("AHI1", "ALMS1")
)

p <- gsHistogram(genesets, gs_names, binwidth = 1)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
p <- gsHistogram(genes, macrophage_topGO_example_small$Genesets)

Calculate clusters based on kMeans clustering

Description

This function performs kMeans clustering on a set of scores.

Usage

kMeansClustering(scores, k, iter = 500, nstart = 50)

Arguments

scores

A Matrix::Matrix() of (distance) scores
k

numerical, the number of centers to start with. This number will correlate with the resulting number of clusters.
iter

numerical, number of iterations for refinement. Defaults to 500.
nstart

numerical, how often the start points should be switched. Ensures a robust clustering, as clustering is influenced by the start points. Defaults to 50.

Value

A list of clusters

Examples

## Mock example showing how the data should look like
scores <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(scores) <- colnames(scores) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
cluster <- kMeansClustering(scores, k = 3)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

cluster <- kMeansClustering(scores_macrophage_topGO_example_small,
                            k = 5)

Calculate clusters based on kNN clustering

Description

This function performs k-Nearest Neighbors (kNN) clustering on a set of scores.

Usage

kNN_clustering(scores, k)

Arguments

scores

A Matrix::Matrix() of (distance) scores
k

numerical, the number of neighbors

Value

A list of clusters

Examples

## Mock example showing how the data should look like
scores <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(scores) <- colnames(scores) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
cluster <- kNN_clustering(scores, k = 3)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

kNN <- kNN_clustering(scores_macrophage_topGO_example_small,
                        k = 5)

Cluster genesets using Louvain clustering.

Description

This function is a wrapper function for the Louvain clustering. The actual computation of the clustering is done in the GeDi::clustering() function. This function is mainly a wrapper function for stand-alone use of GeDi functionalities to enhance user experience and allow for a clearer distinction of the individual clustering algorithms.

Usage

louvainClustering(scores, threshold)

Arguments

scores

A Matrix::Matrix() of (distance) scores
threshold

numerical, A threshold used to determine which genesets are considered similar. Genesets are considered similar if (distance) score <= threshold. similar.

Value

A list of clusters

Examples

## Mock example showing how the data should look like
m <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(m) <- colnames(m) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
louvainCluster <- louvainClustering(m, 0.3)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

louvainCluster <- louvainClustering(scores_macrophage_topGO_example_small,
                        threshold = 0.5)

A sample input RData file

Description

A sample input RData file generated from the macrophage dataset.

Format

A data.frame object

Details

This sample input contains data from the macrophage package found on Bioconductor. The exact steps used to generated this file can be found in the package vignette. The used database for the enrichment was the KEGG database.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

A sample input RData file

Description

A sample input RData file generated from the macrophage dataset.

Format

A data.frame object

Details

This sample input contains data from the macrophage package found on Bioconductor. The exact steps used to generated this file can be found in the package vignette. The used database for the enrichment was the Reactome database.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

A sample input RData file

Description

A sample input RData file generated from the macrophage dataset.

Format

A data.frame object

Details

This sample input contains data from the macrophage package found on Bioconductor. The exact steps used to generated this file can be found in the package vignette.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

A small sample input RData file

Description

A small sample input RData file generated from the macrophage dataset.

Format

A data.frame object

Details

This sample input contains data from the macrophage package found on Bioconductor. It is a small version of the macrophage_topGO_example and only contains the first 50 rows of this example. It can be used for fast testing of the application.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

Cluster genesets using Markov clustering.

Description

This function is a wrapper function for the Markov clustering. The actual computation of the clustering is done in the GeDi::clustering() function. This function is mainly a wrapper function for stand-alone use of GeDi functionalities to enhance user experience and allow for a clearer distinction of the individual clustering algorithms.

Usage

markovClustering(scores, threshold)

Arguments

scores

A Matrix::Matrix() of (distance) scores
threshold

numerical, A threshold used to determine which genesets are considered similar. Genesets are considered similar if (distance) score <= threshold. similar.

Value

A list of clusters

Examples

## Mock example showing how the data should look like
m <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(m) <- colnames(m) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
markovCluster <- markovClustering(m, 0.3)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

markovCluster <- markovClustering(scores_macrophage_topGO_example_small,
                        threshold = 0.5)

Calculate clusters based on PAM clustering

Description

This function performs Partioning aroung Medoids clustering on a set of scores.

Usage

pamClustering(scores, k)

Arguments

scores

A Matrix::Matrix() of (distance) scores
k

numerical, the number of centers to start with. This number will correlate with the resulting number of clusters.

Value

A list of clusters

Examples

## Mock example showing how the data should look like
scores <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
rownames(scores) <- colnames(scores) <- c("a", "b", "c", "d", "e",
                                "f", "g", "h", "i", "j")
cluster <- pamClustering(scores, k = 3)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

cluster <- pamClustering(scores_macrophage_topGO_example_small,
                               k = 5)

Calculate local pMM distance

Description

Calculate the local pMM distance of two genesets.

Usage

pMMlocal(a, b, ppi, maxInteract, alpha)

Arguments

a, b

character vector, set of gene identifiers.
ppi

a data.frame, Protein-protein interaction (PPI) network data frame. The object is expected to have three columns, Gene1 and Gene2 which specify the gene names of the interacting proteins in no particular order (symmetric interaction) and a column combined_score which is a numerical value of the strength of the interaction.
maxInteract

numeric, Maximum interaction value in the PPI.
alpha

numeric, Scaling factor for controlling the influence of the interaction score. Defaults to 1.

Value

The pMMlocal score between the two gene sets.

References

See https://doi.org/10.1186/s12864-019-5738-6 for details on the original implementation.

Examples

## Mock example showing how the data should look like
a <- c("PDHB", "VARS2")
b <- c("IARS2", "PDHA1")

ppi <- data.frame(
  Gene1 = c("PDHB", "VARS2"),
  Gene2 = c("IARS2", "PDHA1"),
  combined_score = c(0.5, 0.2)
)
maxInteract <- max(ppi$combined_score)

pMM_score <- pMMlocal(a, b, ppi, alpha = 1,  maxInteract)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- GeDi::prepareGenesetData(macrophage_topGO_example_small)
data(ppi_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
maxInteract <- max(ppi_macrophage_topGO_example_small$combined_score)

pMMlocal <- pMMlocal(genes[1], genes[2], ppi, alpha = 1,  maxInteract)

PPI

Description

A file containing a Protein-Protein Interaction (PPI) data.frame for the macrophage_topGO_example_small.

Format

A data.frame object

Details

This sample input contains a PPI for the macrophage_topGO_example_small. The PPI has been downloaded using the functions to download a PPI matrix. Please check out the vignette for further information.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

Split string of genes

Description

Split a long string of space separated genes into a list of individual genes.

Usage

prepareGenesetData(genesets, gene_name = NULL)

Arguments

genesets

a data.frame, A data.frame with at least two columns. One should be called Geneset, containing the names/identifiers of the genesets in the data. The second column should be called Genes and contains one string of the genes contained in each geneset.
gene_name

a character, Alternative name for the column containing the genes in genesets. If not given, the column is expected to be called Genes.

Value

A list containing for each geneset in the Geneset column a list of the included genes.

Examples

## Mock example showing how the data should look like
df <- data.frame(
  Geneset = c(
    "Cell Cycle",
    "Biological Process",
    "Mitosis"
  ),
  Genes = c(
    c("PDHB,VARS2,IARS2"),
    c("LARS,LARS2"),
    c("IARS,SUV3")
  )
)
genes <- prepareGenesetData(df)

## Example using the data available in the package
data(macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())
genes <- prepareGenesetData(macrophage_topGO_example_small)

A sample input text file

Description

A sample input text file taken from the GScluster package

Format

Text file

Details

This sample input text file contains data from the GScluster package. It is identical to the sample_geneset.txt file found on the Github page of the package.

References

Yoon, S., Kim, J., Kim, SK. et al. GScluster: network-weighted gene-set clustering analysis. BMC Genomics 20, 352 (2019). https://doi.org/10.1186/s12864-019-5738-6

A broken input text file

Description

A broken input text file to test the application

Format

Text file

Details

This sample input text file is broken and used for testing the application.

An empty input text file

Description

An empty input text file to test the application

Format

Text file

Details

This sample input text file is empty and used for testing the application.

A small sample input text file

Description

A sample input text file taken from the GScluster package, which is reduced to a smaller number of entries for faster testing of the application.

Format

Text file

Details

This sample input text file contains data from the GScluster package. It was taken from the sample_geneset.txt file found on the Github page of the package and then reduced to a smaller amount of entries for faster testing of the application.

References

Yoon, S., Kim, J., Kim, SK. et al. GScluster: network-weighted gene-set clustering analysis. BMC Genomics 20, 352 (2019). https://doi.org/10.1186/s12864-019-5738-6

Scaling (distance) scores

Description

A method to scale a matrix of distance scores with the GO term similarity of the associated genesets.

Usage

scaleGO(
  scores,
  geneset_ids,
  method = "Wang",
  ontology = "BP",
  species = "org.Hs.eg.db",
  BPPARAM = BiocParallel::SerialParam()
)

Arguments

scores

a Matrix::Matrix(), a matrix of (distance) scores for the identifiers in geneset_ids.
geneset_ids

list, a list of GO identifiers to score
method

character, the method to calculate the GO distance. See GOSemSim::goSim measure parameter for possibilities.
ontology

character, the ontology to use. See GOSemSim::goSim ont parameter for possibilities.
species

character, the species of your data. Indicated as org.XX.eg.db package from Bioconductor.
BPPARAM

A BiocParallelParam object specifying how parallelization should be handled

Value

A Matrix::Matrix() of scaled values.

Examples

## Mock example showing how the data should look like
go_ids <- c("GO:0002503", "GO:0045087", "GO:0019886",
            "GO:0002250", "GO:0001916", "GO:0019885")
set.seed(42)
scores <- Matrix::Matrix(stats::runif(36, min = 0, max = 1), 6, 6)
similarity <- scaleGO(scores,
                      go_ids)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small, package = "GeDi")
data(macrophage_topGO_example_small, package = "GeDi")
go_ids <- macrophage_topGO_example_small$Genesets
## Not run: 
scores_scaled <- scaleGO(scores_macrophage_topGO_example_small,
                         go_ids)

## End(Not run)

Sample scores

Description

A file containing sample distance scores for the macrophage_topGO_example_small.

Format

A sparse matrix (dgCMatrix)

Details

This sample input contains scores for the macrophage_topGO_example_small. Distance scores have been calculated using the getJaccardMatrix() method.

References

Alasoo, K., Rodrigues, J., Mukhopadhyay, S. et al. Shared genetic effects on chromatin and gene expression indicate a role for enhancer priming in immune response. Nat Genet 50, 424–431 (2018). https://doi.org/10.1038/s41588-018-0046-7

Find clustering seeds

Description

Determine initial seeds for the clustering from the distance score matrix.

Usage

seedFinding(distances, simThreshold, memThreshold)

Arguments

distances

A Matrix::Matrix() of (distance) scores
simThreshold

numerical, A threshold to determine which genesets are considered close (i.e. have a distance <= simThreshold) in the distances matrix.
memThreshold

numerical, A threshold used to ensure that enough members of a potential seed set are close/similar to each other. Only if this condition is met, the set is considered a seed.

Value

A list of seeds which can be used for clustering

References

See https://david.ncifcrf.gov/helps/functional_classification.html#clustering for details on the original implementation

Examples

## Mock example showing how the data should look like

m <- Matrix::Matrix(stats::runif(100, min = 0, max = 1), 10, 10)
seeds <- seedFinding(distances = m, simThreshold = 0.3, memThreshold = 0.5)

## Example using the data available in the package
data(scores_macrophage_topGO_example_small,
     package = "GeDi",
     envir = environment())

seeds <- seedFinding(scores_macrophage_topGO_example_small,
                     simThreshold = 0.3,
                     memThreshold = 0.5)