gprofiler2 -- an R package for gene list functional enrichment analysis and namespace conversion toolset g:Profiler.
Kolberg L, Raudvere U, Kuzmin I, Vilo J, Peterson H. · 2020
Read gprofiler2 -- an R package for gene list functional enrichment analysis and namespace conversion toolset g:Profiler. by Kolberg L, Raudvere U, Kuzmin I, Vilo J, Peterson H. (2020) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/europmc-pmc7859841.txt.
Introduction
Kolberg and co-authors present gprofiler2, an R package for gene list analysis. The g:Profiler web toolset has served gene studies since 2007. It profiles what gene lists do and converts ID types. Lab teams wanted that power inside R scripts.
The package reaches g:Profiler through its web API. Enrichment tests run inside automated pipelines. Plot tools help read and publish the results. Users can map IDs between hundreds of types. Ortholog species lookups come along too. The package is free on the CRAN archive. It aims to make gene list work repeatable and scriptable.
Catalog record
| CREATOR | Kolberg L, Raudvere U, Kuzmin I, Vilo J, Peterson H. |
|---|---|
| YEAR | 2020 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 43 KB |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
gprofiler2 -- an R package for gene list functional enrichment analysis and namespace conversion toolset g:Profiler By Liis Kolberg, Uku Raudvere, Ivan Kuzmin, Jaak Vilo, Hedi Peterson Published in: F1000Research DOI: 10.12688/f1000research.24956.2 ABSTRACT Abstract g:Profiler ( https://biit.cs.ut.ee/gprofiler) is a widely used gene list functional profiling and namespace conversion toolset that has been contributing to reproducible biological data analysis already since 2007. Here we introduce the accompanying R package, gprofiler2, developed to facilitate programmatic access to g:Profiler computations and databases via REST API. The gprofiler2 package provides an easy-to-use functionality that enables researchers to incorporate functional enrichment analysis into automated analysis pipelines written in R. The package also implements interactive visualisation methods to help to interpret the enrichment results and to illustrate them for publications. In addition, gprofiler2 gives access to the versatile gene/protein identifier conversion functionality in g:Profiler enabling to map between hundreds of different identifier types or orthologous species. The gprofiler2 package is freely available at the CRAN repository. Keywords: g:Profiler, R package, functional enrichment analysis, identifier mapping, Gene Ontology, pathways INTRODUCTION Interpretation of gene lists is a key step in numerous biological data analysis workflows, such as differential gene expression analysis and co-expression clustering of RNA-seq or microarray data. Usually this involves associating these gene lists with previous knowledge from well curated data sources of biological processes and pathways. However, as the knowledge bases are constantly changing, keeping the associations up to date requires careful data management. Handling numerous databases, especially when using different gene identifier types, can be a very time-consuming process for researchers. g:Profiler ( https://biit.cs.ut.ee/gprofiler) is a popular web toolset that helps to handle gene lists from various biological and biomedical studies of more than 600 species and strains, including vertebrates, plants, fungi, insects and parasites 1, 2. g:Profiler’s best known functionality is the over-representation analysis to identify significantly enriched biological functions and pathways obtained from well established data sources which include, among others, Gene Ontology (GO) 3, KEGG 4 and Reactome 5. The information about genes, identifier types and GO term associations in g:Profiler is mostly based on Ensembl databases 6 including data from Ensembl Genomes, fungi, plants and metazoa specific versions of Ensembl. g:Profiler follows Ensembl’s quarterly update cycle while keeping the access to previous data versions as archives for reproducibility. The parasite specific data is included from WormBase 7. Providing users with fast and easy access has been the main goal of g:Profiler developers. Since 2007, g:Profiler has been in constant development and with the recent update in 2019 a new accompanying R package, gprofiler2, was developed 8. The R package relies on the g:Profiler REST API requests providing an easy programmatic access to the same functionalities as in the web tool without performing heavy computations and mappings in R. While there are other popular R packages for functional enrichment analysis, such as topGO 9 and clusterProfiler 10, gprofiler2 provides access to numerous annotation data sources with a single query without requiring to download any of these sources to a local computer. Furthermore, the mapping between different gene identifiers is automatic and the input can be a mixed list of identifiers. g:Profiler’s continuous development and flexibility of usage has been recognised by the European Life Science Infrastructure ELIXIR, which has selected it as one of its Recommended Interoperability Resources. g:Profiler development team encourages and supports external tools and packages to use either gprofiler2 package or the public API to be part of their workflows. For example, RCAS Bioconductor package 11 includes gprofiler2 for functional analysis of transcriptomic regions detected by different high-throughput experiments. Single-cell mapper package ( scMappR) 12 analyses cell-type specific gene lists with gprofiler2. OmnipathR 13 suggests using gprofiler2 for enrichment analysis of protein complexes. Gene Co-expression Network analysis pipeline ( GWENA) uses gprofiler2 in their pipeline for functional enrichment of co-expressed gene modules. A Nextflow differential gene expression analysis pipeline includes gprofiler2 for pathway analysis. Here we demonstrate how to conveniently incorporate the gprofiler2 R package into bioinformatics analysis pipelines using differential gene expression analysis as an example. METHODS Implementation Inherently, gprofiler2 8 is a collection of wrapper functions in R that simplify sending POST requests to the g:Profiler REST API using the RCurl package 14. This means that all the annotation data sources and computations are centralised in a single well-maintained server and therefore the results from both the web tool and R package are guaranteed to be identical. Relying on the central API also simplifies the maintenance of the g:Profiler interfaces and enables the R users to get access to the most up-to-date data without having to download the heavy annotation data files to their own devices. At the same time, g:Profiler respects users’ privacy and does not store input gene lists unless the user explicitly requests their data to be stored for future reference via dedicated short links (see section “Sending analysis from R to g:Profiler web interface”). There are four main API wrapper functions in gprofiler2: - gost for functional enrichment analysis - gconvert for mapping gene identifiers between different namespaces - gorth for mapping orthologous genes across species - gsnpense for mapping SNP rs-IDs to chromosome positions, genes and variant effects. In addition to fetching the results from the API, gprofiler2 uses the packages ggplot2 15 and plotly 16 to provide visualisations for enrichment results that are similar to the web tool ones. Using ggplot2 allows users to customise the visualisations by adding or removing graphical layers, and to adjust the quality of images for publication. This article was written using R version 3.6.1 (2019-07-05) and gprofiler2 version 0.1.9. Operation The gprofiler2 R package is available from CRAN and works on R versions 3.5 and above. The package also includes a detailed vignette. The package can be installed from CRAN: # install from CRAN install.packages("gprofiler2") # load the package library(gprofiler2) Input description The most popular functionality of g:Profiler is functional enrichment analysis provided by the g:GOSt tool that performs over-representation analysis using hypergeometric test. This functionality is available in gprofiler2 under the function gost. The required inputs for this function are a vector of gene identifiers, query, and the name of the corresponding organism which is constructed by concatenating the first letter of the genus name and the specific epithet, e.g. hsapiens for human genes. The full list of supported species and strains, 641 in total, is available on the g:Profiler web page. The query vector can include mixed types of gene/protein identifiers, SNP rs-IDs, chromosomal intervals or term IDs. Accepting a mixture of IDs is a unique feature that skips time-consuming manual steps of converting between different identifier types required by other functional enrichment tools. However, in case of analysing numeric identifiers (e.g. Entrez IDs) the user should specify the namespace using the numeric_ns parameter. The same description of input query and organism holds for the three other functions in gprofiler2. gostres = gost(query = c("X:1000:1000000", "rs17396340", "GO:0005005", "ENSG00000156103", "NLRP1", "3837"), organism = "hsapiens", numeric_ns = "ENTREZGENE_ACC") Several additional parameters in the gost function help to perform the analysis according to specific needs, including custom statistical options such as background definition, statistical significance threshold, method for multiple testing correction and testing for under-representation. Also, additional information like GO evidence codes and genes belonging