In waldronlab/MultiAssayExperimentData: Exposes and Makes Available Data from the cBioPortal Web Resources

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  cache = TRUE,
  out.width = "100%"
)
options(tibble.print_min = 5, tibble.print_max = 5)
library(BiocStyle)

cBioPortalData

cBioPortal data and MultiAssayExperiment

Overview

The cBioPortalData R package aims to import cBioPortal datasets as r Biocpkg("MultiAssayExperiment") objects into Bioconductor. Some of the features of the package include:

1. The use of the MultiAssayExperiment integrative container for coordinating and representing the data.
2. The data container explicitly links all assays to the patient clinical/pathological data.
3. With a flexible API, MultiAssayExperiment provides harmonized subsetting and reshaping into convenient wide and long formats.
4. The package provides datasets from both the API and the saved packaged data.
5. It also provides automatic local caching, thanks to r Biocpkg("BiocFileCache")

MultiAssayExperiment Cheatsheet

Quick Start

Installation

To install from Bioconductor (recommended for most users, this will install the release or development version corresponding to your version of Bioconductor):

if (!require("BiocManager", quietly = TRUE))
    install.packages("BiocManager")

BiocManager::install("cBioPortalData")

Developers may want to install from GitHub for bleeding-edge updates (although this is generally not necessary because changes here are also pushed to bioc-devel). Note that developers must be working with the development version of Bioconductor; see bioc-devel for details.

if (!require("cBioPortalData", quietly = TRUE))
    BiocManager::install("waldronlab/cBioPortalData")

To load the package:

library(cBioPortalData)

cbio <- cBioPortal()
studies <- getStudies(cbio, buildReport = TRUE)
st <- studies
packpct <- round(prop.table(table(st$pack_build))[[2]], 2) * 100
apipct <- round(prop.table(table(st$api_build))[[2]], 2) * 100

Note

cBioPortalData is a work in progress due to changes in data curation and cBioPortal API specification. Users can view the data(studiesTable) dataset to get an overview of the studies that are available and currently building as MultiAssayExperiment representations. About r apipct % of the studies via the API (api_build) and r packpct % of the package studies (pack_build) are building, these include additional datasets that were not previously available. Feel free to file an issue to request prioritization of fixing any of the remaining datasets.

cbio <- cBioPortal()
studies <- getStudies(cbio, buildReport = TRUE)

table(studies$api_build)

table(studies$pack_build)

API Service

Flexible and granular access to cBioPortal data from cbioportal.org/api. This option is best used with a particular gene panel of interest. It allows users to download sections of the data with molecular profile and gene panel combinations within a study.

gbm <- cBioPortalData(api = cbio, by = "hugoGeneSymbol", studyId = "gbm_tcga",
    genePanelId = "IMPACT341",
    molecularProfileIds = c("gbm_tcga_rppa", "gbm_tcga_mrna")
)

gbm

Packaged Data Service

This function will download a dataset from the cbioportal.org/datasets website as a packaged tarball and serve it to users as a MultiAssayExperiment object. This option is good for users who are interested in obtaining all the data for a particular study.

acc <- cBioDataPack("acc_tcga")

acc

waldronlab/MultiAssayExperimentData documentation built on May 4, 2024, 2:29 p.m.