| 1 | 1 |
new file mode 100644 |
| ... | ... |
@@ -0,0 +1,13 @@ |
| 1 |
+* using log directory ‘/home/veit/devel/Bioinformatics/DataAnalysis/ShinyApps/vsclust/..Rcheck’ |
|
| 2 |
+* using R version 4.4.2 (2024-10-31) |
|
| 3 |
+* using platform: x86_64-pc-linux-gnu |
|
| 4 |
+* R was compiled by |
|
| 5 |
+ gcc (Ubuntu 13.2.0-23ubuntu4) 13.2.0 |
|
| 6 |
+ GNU Fortran (Ubuntu 13.2.0-23ubuntu4) 13.2.0 |
|
| 7 |
+* running under: Ubuntu 24.04.1 LTS |
|
| 8 |
+* using session charset: UTF-8 |
|
| 9 |
+* checking for file ‘./DESCRIPTION’ ... ERROR |
|
| 10 |
+Required fields missing or empty: |
|
| 11 |
+ ‘Author’ ‘Maintainer’ |
|
| 12 |
+* DONE |
|
| 13 |
+Status: 1 ERROR |
| ... | ... |
@@ -80,9 +80,9 @@ shinyUI(fluidPage(theme=shinytheme("cosmo"),
|
| 80 | 80 |
h3("Distribution of features over clusters"),
|
| 81 | 81 |
DT::dataTableOutput("clustinf2"),
|
| 82 | 82 |
value="clust2"), |
| 83 |
- tabPanel("Enriched terms (DAVID)",inputPanel(
|
|
| 83 |
+ tabPanel("Enriched terms (STRING)",inputPanel(
|
|
| 84 | 84 |
# radioButtons("enrich_method","Enrichment tool",choices=c("DAVID"="DAVID","KEGG (local copy)"="KEGG","GO molecular function (local copy)"="GOMF"),selected = "DAVID"),br(),
|
| 85 |
- selectInput("infosource","Information resource (DAVID)",choices=list("GO terms"=c("GO molecular function"="Function",
|
|
| 85 |
+ selectInput("infosource","Information resource (STRING)",choices=list("GO terms"=c("GO molecular function"="Function",
|
|
| 86 | 86 |
"GO biological process"="Process", |
| 87 | 87 |
"GO cellular component"="Component"), |
| 88 | 88 |
UniProt = c("UniProt keywords" = "Keyword"),
|
| ... | ... |
@@ -2,7 +2,7 @@ |
| 2 | 2 |
% Please edit documentation in R/WrapperFuncs.R |
| 3 | 3 |
\name{enrichSTRING_API}
|
| 4 | 4 |
\alias{enrichSTRING_API}
|
| 5 |
-\title{Functional Enrichment with STRING}
|
|
| 5 |
+\title{Enrichment Analysis via STRING REST API}
|
|
| 6 | 6 |
\usage{
|
| 7 | 7 |
enrichSTRING_API( |
| 8 | 8 |
genes, |
| ... | ... |
@@ -13,57 +13,75 @@ enrichSTRING_API( |
| 13 | 13 |
) |
| 14 | 14 |
} |
| 15 | 15 |
\arguments{
|
| 16 |
-\item{cl}{A clustering result. Typically either the direct output of
|
|
| 17 |
-\code{vsclust_algorithm} or a \code{Bestcl} object from \code{ClustComp}
|
|
| 18 |
-or \code{runClustWrapper}.}
|
|
| 16 |
+\item{genes}{Character vector of gene or protein identifiers recognized by STRING.}
|
|
| 19 | 17 |
|
| 20 |
-\item{protnames}{Optional named vector mapping feature IDs (as in \code{cl$cluster})
|
|
| 21 |
-to more interpretable gene/protein identifiers. If \code{NULL} (the default),
|
|
| 22 |
-the feature names in \code{cl} themselves are used.}
|
|
| 18 |
+\item{species}{A single numeric or string specifying the NCBI taxonomy ID
|
|
| 19 |
+(e.g., \code{9606} for human). If \code{"none"} (default), no species is set in
|
|
| 20 |
+the request, and STRING will attempt to auto-detect or may fail.} |
|
| 23 | 21 |
|
| 24 |
-\item{infosource}{Currently unused; previously indicated enrichment categories
|
|
| 25 |
-when using DAVID. Kept for compatibility but is ignored in this version.} |
|
| 22 |
+\item{category}{One or more enrichment categories from STRING (e.g., \code{"KEGG"},
|
|
| 23 |
+\code{"Process"}, etc.). Defaults to \code{"KEGG"} if unspecified.}
|
|
| 24 |
+ |
|
| 25 |
+\item{adjpvalueCutoff}{Numeric cutoff for the BH-adjusted p-value (default 0.05).}
|
|
| 26 |
+ |
|
| 27 |
+\item{verbose}{Logical indicating whether to print diagnostic messages (default \code{FALSE}).}
|
|
| 26 | 28 |
} |
| 27 | 29 |
\value{
|
| 28 |
-A list with three components: |
|
| 29 |
-\describe{
|
|
| 30 |
- \item{\code{fullFuncs}}{The full \code{compareCluster} result from \code{enrichSTRING_API}.}
|
|
| 31 |
- \item{\code{redFuncs}}{A \code{compareClusterResult} object containing only
|
|
| 32 |
- the top 20 enriched terms (if more than 20 were detected).} |
|
| 33 |
- \item{\code{BHI}}{A numeric value from \code{calcBHI} measuring cluster
|
|
| 34 |
- heterogeneity.} |
|
| 35 |
-} |
|
| 30 |
+If any terms pass the cutoff, an \code{\link{enrichResult}} object
|
|
| 31 |
+ (from \emph{clusterProfiler}) is returned. Otherwise, if \emph{clusterProfiler}
|
|
| 32 |
+ is not installed or no terms pass filtering, the function either returns a |
|
| 33 |
+ data.frame (if installed but no terms pass) or \code{NULL}.
|
|
| 36 | 34 |
} |
| 37 | 35 |
\description{
|
| 38 |
-\code{runFuncEnrich} performs a functional enrichment analysis for each cluster
|
|
| 39 |
-of features (genes/proteins) using an \code{enrichSTRING_API} call (which queries
|
|
| 40 |
-the STRING database). It replaces a previously used approach that relied on |
|
| 41 |
-\emph{RDAVIDWebService}, and is therefore more up to date.
|
|
| 36 |
+\code{enrichSTRING_API} performs a functional enrichment analysis by sending a
|
|
| 37 |
+gene list to STRING's TSV API endpoint, retrieving enrichment results for one or |
|
| 38 |
+more categories (e.g., "KEGG"), and building a \code{clusterProfiler}-style result
|
|
| 39 |
+object. It is intended as a lightweight replacement for older web-service–based |
|
| 40 |
+methods like \emph{RDAVIDWebService}.
|
|
| 42 | 41 |
} |
| 43 | 42 |
\details{
|
| 44 |
-The function takes a clustering result (e.g., from \code{vsclust_algorithm} or
|
|
| 45 |
-\code{Bestcl} objects) and, for each cluster:
|
|
| 43 |
+This function: |
|
| 46 | 44 |
\enumerate{
|
| 47 |
- \item Extracts the member features with membership above 0.5. |
|
| 48 |
- \item Optionally replaces their IDs with entries from \code{protnames}.
|
|
| 49 |
- \item Calls \code{compareCluster} (from \emph{clusterProfiler}) using a
|
|
| 50 |
- custom \code{enrichSTRING_API} function for the actual STRING-based enrichment.
|
|
| 45 |
+ \item Accepts a vector of \code{genes} recognized by STRING (e.g., Ensembl,
|
|
| 46 |
+ UniProt, or commonly used gene symbols for a given species). |
|
| 47 |
+ \item Sends a \emph{POST} request to \url{https://string-db.org/api/tsv/enrichment} with
|
|
| 48 |
+ those identifiers (and optionally \code{species}).
|
|
| 49 |
+ \item Parses the returned \emph{TSV}-formatted enrichment data, which typically
|
|
| 50 |
+ includes columns for \code{term}, \code{description}, \code{p_value}, and more.
|
|
| 51 |
+ \item Optionally filters by \code{category} (e.g., "KEGG", "Process", etc.), applies
|
|
| 52 |
+ \code{BH} multiple-testing correction, and removes terms above the
|
|
| 53 |
+ \code{adjpvalueCutoff}.
|
|
| 54 |
+ \item Returns an object of class \code{enrichResult} if \emph{clusterProfiler} is
|
|
| 55 |
+ installed; otherwise, it simply returns the filtered data frame. |
|
| 51 | 56 |
} |
| 52 |
-The resulting \code{compareClusterResult} includes adjusted p-values (FDR),
|
|
| 53 |
-and the top 20 enriched terms are retained if the total set is larger than 20. |
|
| 57 |
+ |
|
| 58 |
+Note that this function does \emph{not} allow you to explicitly supply a custom
|
|
| 59 |
+background gene set. The STRING API by default uses the entire known set of genes |
|
| 60 |
+or proteins for the specified species as the background. Also, the total number of |
|
| 61 |
+background genes is not reported by STRING; only how many of them map to each term. |
|
| 54 | 62 |
} |
| 55 | 63 |
\examples{
|
| 56 |
-\dontrun{
|
|
| 57 |
- # Suppose 'mycl' is a clustering result from vsclust_algorithm, |
|
| 58 |
- # and we have a named vector 'myProtnames' that maps feature IDs to gene symbols: |
|
| 64 |
+ library(httr) |
|
| 59 | 65 |
|
| 60 |
- res <- runFuncEnrich(mycl, protnames = myProtnames) |
|
| 61 |
- res$fullFuncs # The full compareCluster output |
|
| 62 |
- res$redFuncs # The top 20 enriched terms |
|
| 63 |
- res$BHI # The numeric BHI value |
|
| 64 |
-} |
|
| 66 |
+ # A small gene set: |
|
| 67 |
+ gene_set <- c("TP53","BRCA1","BRCA2","EGFR")
|
|
| 68 |
+ |
|
| 69 |
+ # Perform enrichment on KEGG terms for human (9606): |
|
| 70 |
+ enr <- enrichSTRING_API( |
|
| 71 |
+ genes = gene_set, |
|
| 72 |
+ species = 9606, |
|
| 73 |
+ category = "KEGG", |
|
| 74 |
+ adjpvalueCutoff = 0.05, |
|
| 75 |
+ verbose = TRUE |
|
| 76 |
+ ) |
|
| 77 |
+ |
|
| 78 |
+ if (!is.null(enr)) {
|
|
| 79 |
+ # If clusterProfiler is installed and some terms pass filtering, check results: |
|
| 80 |
+ head(enr@result) |
|
| 81 |
+ } |
|
| 65 | 82 |
|
| 66 | 83 |
} |
| 67 | 84 |
\seealso{
|
| 68 |
-\code{\link{compareCluster}}, \code{\link{enrichSTRING_API}}
|
|
| 85 |
+\href{https://string-db.org/help/api/}{STRING API documentation},
|
|
| 86 |
+\code{\link{clusterProfiler}}.
|
|
| 69 | 87 |
} |
