NEWS

cBioPortalData 2.16.0

Add cgdsr to cBioPortalData migration vignette (@kmezhoud, #54)
Unmapped experiments are now added to the metadata in cBioDataPack
Set default api. = api/v2/api-docs in cBioPortal to access the API protocol's new location
The fetchData developer function added to handle both molecularData and mutationData requests as deduced from the molecularProfileIds vector

Check for valid studyIds with getStudies in cBioDataPack
ask argument correctly passed down to caching mechanism in cBioDataPack
check_build option available in cBioDataPack particularly for new studies that have not been checked against.

Auth token string or file can now be included in the cBioPortal function.
The check_build argument can be set to FALSE for alternative APIs, e.g., KidsFirst, when using cBioPortalData
queryGeneTable translates gene IDs ('hugoGeneSymbols' <> 'entrezGeneIds') via the API service
getDataByGenes supersedes getDataByGenePanel
getStudies() replaces data('studiesTable') to discover study IDs

Fixed issue where the by argument was not passed to getDataByGenes in internal calls
Add names to metadata elements that originate from GISTIC datasets.

A study's build status can be obtained from getStudies(), which has replaced data('studiesTable').
Partial loading of data files supported. A warning is emitted when a data file is not able to be loaded in cBioDataPack.
cBioPortalData checks the data(studiesTable) to verify that study datasets are building, otherwise provide a message in interactive sessions.

Vignettes include additional information (#38, @lwaldron)
getDataByGenePanel deprecated for getDataByGenes which handles input of both gene panels and genes
cBioPortalData now allows for gene inputs as either Entrez IDs or Hugo symbols (#24, @jucor) and sampleIds input
When gene inputs are provided, the by argument has to agree with the type of genes provided (either be entrezGeneId or hugoGeneSymbol).

Fixed an issue where the labels in the metadata from cBioDataPack were missing ('LICENSE' and 'Fusion'; #37)
loadStudy allows cleanup=TRUE for removing files after untar-ing
Published article now available with citation("cBioPortalData")

studiesTable includes additional columns pack_build and api_build to indicate to the user which datasets have been successfully built as MultiAssayExperiment objects. Users will be notified when a dataset, reported as not building, is requested from the cBioDataPack function.
Add sampleIds argument to getDataByGenePanel as part of cache re-work
Allow more flexibility in the hostname when accessing the API with cBioPortal (@inodb, #16)
cBioDataPack downloads from a more robust repository (AWS S3; @inodb, #22)
removePackCache and removeDataCache now remove data from the user's cache based on inputs to respective functions (cBioDataPack and cBioPortalData)

Attempt to merge additional clinical data files from tarballs in cBioDataPack.
Switch to using read.delim instead of read_tsv internally to avoid assigning NA to chromosome column
Use 'PATIENT_ID' when available to determine if experiment data is provided in the tarball files.
Add tests using testthat
Update and include percentages of studies successfully imported using cBioDataPack and cBioPortalData in the documentation
Fix read-in when identifiers are numeric instead of character (@jucor, #27)
Include pagination parameters in geneTable function (@xinwei-sher, #29)

Bioconductor release!
Updated the README.md file from R Markdown file.
Uses the latest version of rapiclient on CRAN
Prepare package for Bioconductor submission
Include protein metadata as a RaggedExperiment from mutation molecular profiles (TCGA only)

Package supports nearly all study identifiers based on recent tests
Only a handful of study identifiers are unsuccessful (create an issue to prioritize).

Make better use of the API return values to craft the sample map for MultiAssayExperiment creation
Additional data included in the metadata slot of the MultiAssayExperiment object. Future revisions will include this data as rowData.
Change vignette titles for build

cBioDataPack allows users to download packaged data objects from download.cbioportal.org/
Data packs are cached using BiocFileCache to avoid re-downloading
cBioPortalData lets users query the cbioportal.org API and retrieve slices of data according to gene, molecular profile identifiers, etc.
Queries through cBioPortalData use a caching mechanism to avoid repeat downloads of data and improve load times
Both functions return a MultiAssayExperiment as the primary data representation
Only a number of study datasets are currently possible to load. Issues can arise with mismatched or munged identifiers
The cBioPortal API representation is handled by the AnVIL package which makes use of rapiclient to provide an automatic R interface to the API

Data pack downloads use an alternative method for download when a download fails