--- title: "Creating An ExperimentHub Package" author: "Valerie Obenchain and Lori Shepherd" date: "Modified: Aug 2019. Compiled: `r format(Sys.Date(), '%d %b %Y')`" output: BiocStyle::html_document: toc: true toc_depth: 2 vignette: > %\VignetteIndexEntry{Creating An ExperimentHub Package} %\VignetteEngine{knitr::rmarkdown} --- # Overview `ExperimentHubData` provides tools to add or modify resources in Bioconductor's `ExperimentHub`. This 'hub' houses curated data from courses, publications or experiments. The resources are generally not files of raw data (as can be the case in `AnnotationHub`) but instead are `R` / `Bioconductor` objects such as GRanges, SummarizedExperiment, data.frame etc. Each resource has associated metadata that can be searched through the `ExperimentHub` client interface. # New resources Resources are contributed to `ExperimentHub` in the form of a package. The package contains the resource metadata, man pages, vignette and any supporting `R` functions the author wants to provide. This is a similar design to the existing `Bioconductor` experimental data packages except the data are stored in AWS S3 buckets or publicly accessibly sites instead of the data/ directory of the package. Below are the steps required for adding new resources. ## Notify `Bioconductor` team member The man page and vignette examples in the data experiment package will not work until the data are available in `ExperimentHub`. Adding the data to AWS S3 and the metadata to the production database involves assistance from a `Bioconductor` team member. The metadata.csv file will have to be created before the data can officially be added to the hub (See inst/extdata section below). Please read the section of "Storage of Data Files". ## Building the data experiment package When a resource is downloaded from `ExperimentHub` the associated data experiment package is loaded in the workspace making the man pages and vignettes readily available. Because documentation plays an important role in understanding these curated resources please take the time to develop clear man pages and a detailed vignette. These documents provide essential background to the user and guide appropriate use the of resources. Below is an outline of package organization. The files listed are required unless otherwise stated. ### `inst/extdata/` - `metadata.csv`: This file contains the metadata in the format of one row per resource to be added to the `ExperimentHub` database. The file should be generated from the code in inst/scripts/make-metadata.R where the final data are written out with `write.csv(..., row.names=FALSE)`. The required column names and data types are specified in `ExperimentHubData::makeExperimentHubMetadata` See ?`ExperimentHubData::makeExperimentHubMetadata` for details. An example data experiment package metadata.csv file can be found [here](https://github.com/Bioconductor/GSE62944/blob/master/inst/extdata/metadata.csv) ### `inst/scripts/` - `make-data.R`: A script describing the steps involved in making the data object(s). This includes where the original data were downloaded from, pre-processing, and how the final R object was made. Include a description of any steps performed outside of `R` with third party software. It is encouraged to serialize Data objects with `save()` with the .rda extension on the filename but not strictly necessary. If the data is provided in another format an appropriate loading method may need to be implemented. Please advise when reaching out for "Uploading Data to S3". - `make-metadata.R`: A script to make the metadata.csv file located in inst/extdata of the package. See ?`ExperimentHubData::makeExperimentHubMetadata` for a description of expected fields and data types. `ExperimentHubData::makeExperimentHubMetadata()` can be used to validate the metadata.csv file before submitting the package. ### `vignettes/` - One or more vignettes describing analysis workflows. ### `R/` - `zzz.R`: Optional. You can include a `.onLoad()` function in a zzz.R file that exports each resource name (i.e., title) into a function. This allows the data to be loaded by name, e.g., `resouce123()`. ```{r, eval=FALSE} .onLoad <- function(libname, pkgname) { fl <- system.file("extdata", "metadata.csv", package=pkgname) titles <- read.csv(fl, stringsAsFactors=FALSE)$Title createHubAccessors(pkgname, titles) } ``` `ExperimentHub::createHubAccessors()` and `ExperimentHub:::.hubAccessorFactory()` provide internal detail. The resource-named function has a single 'metadata' argument. When metadata=TRUE, the metadata are loaded (equivalent to single-bracket method on an ExperimentHub object) and when FALSE the full resource is loaded (equivalent to double-bracket method). - `R/*.R`: Optional. Functions to enhance data exploration. ### `man/` - package man page: The package man page serves as a landing point and should briefly describe all resources associated with the package. There should be an \alias entry for each resource title either on the package man page or individual man pages. - resource man pages: Resources can be documented on the same page, grouped by common type or have their own dedicated man pages. - document how data are loaded: Data can be accessed via the standard ExperimentHub interface with single and double-bracket methods, e.g., ```{r, eval=FALSE} library(ExperimentHub) eh <- ExperimentHub() myfiles <- query(eh, "PACKAGENAME") myfiles[[1]] ## load the first resource in the list myfiles[["EH123"]] ## load by EH id ``` - If a `.onLoad()` function is used to export each resource as a function also document that method of loading, e.g., ```{r, eval=FALSE} resourceA(metadata = FALSE) ## data are loaded resourceA(metadata = TRUE) ## metadata are displayed ``` ### `DESCRIPTION` / `NAMESPACE` - The package should depend on and fully import ExperimentHub. If using the suggested `.onLoad()` function, import the utils package in the DESCRIPTION file and selectively importFrom(utils, read.csv) in the NAMESPACE. - Package authors are encouraged to use the `ExperimentHub::listResources()` and `ExperimentHub::loadResource()` functions in their man pages and vignette. These helpers are designed to facilitate data discovery within a specific package vs within all of ExperimentHub. - The biocViews should contain terms from [ExperimentData](http://bioconductor.org/packages/release/BiocViews.html#___ExperimentData) and should also contain the term `ExperimentHub`. ## Data objects Data are not formally part of the software package and are stored separately in a publicly accessible hosted site or by Bioconductor in an AWS S3 buckets. The author should read the following section on "Storage of Data Files". ## Metadata When you are satisfied with the representation of your resources in make-metadata.R (which produces metadata.csv) the `Bioconductor` team member will add the metadata to the production database. Confirm the data in inst/exdata/metadata.csv are valid by running ExperimentHubData:::makeExperimentHubMetadata() on your package. Please address and warnings or errors. ## Package review Once the data are in AWS S3 or public site and the metadata have been added to the production database the man pages and vignette can be finalized. When the package passes R CMD build and check it can be submitted to the [package tracker](https://github.com/Bioconductor/Contributions) for review. The package should be submitted without any of the data that is now located remotely; This keeps the package light weight and minimual size while still providing access to key large data files now stored remotely If the data files were added to the github repository please see [removing large data files and clean git tree](http://bioconductor.org/developers/how-to/git/remove-large-data/) to remove the large files and reduce package size. Many times these data package are created as a suppliment to a software package. There is a process for submitting [mulitple package under the same issue](https://github.com/Bioconductor/Contributions#submitting-related-packages). # Add additional resources Metadata for new versions of the data can be added to the same package as they become available. * The titles for the new versions must be unique and not match the title of any resource currently in ExperimentHub. Good practice would be to include the version and / or genome build in the title. * Make data available: either on publicly accessible site or see section on "Uploading Data to S3" * Update make-metadata.R with the new metadata information * Generate a new metadata.csv file. The package should contain metadata for all versions of the data in ExperimentHub so the old file should remain. When adding a new version it might be helpful to write a new csv file named by version, e.g., metadata_v84.csv, metadata_85.csv etc. * Bump package version and commit to git * Notify hubs@bioconductor.org that an update is ready and a team member will add the new metadata to the production database; new resources will not be visible in ExperimentHub until the metadata are added to the database. Contact hubs@bioconductor.org or maintainer@bioconductor.org with any questions. # Bug fixes A bug fix may involve a change to the metadata, data resource or both. ## Update the resource * The replacement resource must have the same name as the original and be at the same location (path) * Notify hubs@bioconductor.org that you want to replace the data and make the files available: see section "Uploading Data to S3". ## Update the metadata New metadata records can be added for new resources but modifying existing records is discouraged. Record modification will only be done in the case of bug fixes. * Notify hubs@bioconductor.org that you want to change the metadata * Update make-metadata.R with modified information and regenerate the metadata.csv file if necessary * Bump the package version and commit to git # Remove resources Removing resources should be done with caution. The intent is that ExperimentHub be a 'reproducible' resource by providing a stable snapshot of the data. Data made available in Bioconductor version x.y.z should be available for all versions greater than x.y.z. Unfortunately this is not always possible. If you find it necessary to remove data from ExperimentHub please contact hubs@bioconductor.org or maintainer@bioconductor.org for assistance. When a resource is removed from ExperimentHub the 'status' field in the metadata is modified to explain why they are no longer available. Once this status is changed the `ExperimentHub()` constructor will not list the resource among the available ids. An attempt to extract the resource with '[[' and the EH id will return an error along with the status message. The function `getInfoOnIds` will display metadata information for any resource including resources still in the database but no longer available. # Storage of Data Files The data should not be included in the package. This keeps the package light weight and quick for a user to install. This allows the user to investigate functions and documentation without downloading large data files and only proceeding with the download when necessary. There are two options for storing data: Bioconductor AWS S3 buckets or hosting the data elsewhere on a publicly accessible site. See information below and choose the options that fits best for your situation. ## Hosting Data on a Publicly Accessible Site Data can be accessed through the hubs from any publicly accessible site. The metadata.csv file[s] created will need the column `Location_Prefix` to indicate the hosted site. See more in the description of the metadata columns/fields below but as a quick example if the link to the data file is `ftp://mylocalserver/singlecellExperiments/dataSet1.Rds` an example breakdown of the `Location_Prefix` and `RDataPath` for this entry in the metadata.csv file would be `ftp://mylocalserver/` for the `Location_Prefix` and `singlecellExperiments/dataSet1.Rds` for the `RDataPath`. ## Uploading Data to S3 Instead of providing the data files via dropbox, ftp, etc. we will grant temporary access to an S3 bucket where you can upload your data. Please email hubs@bioconductor.org for access. You will be given access to the 'AnnotationContributor' user. Ensure that the `AWS CLI` is installed on your machine. See instructions for installing `AWS CLI` [here](https://aws.amazon.com/cli/). Once you have requested access you will be emailed a set of keys. There are two options to set the profile up for AnnotationContributor 1. Update your `.aws/config` file to include the following updating the keys accordingly: ``` [profile AnnotationContributor] output = text region = us-east-1 aws_access_key_id = **** aws_secret_access_key = **** ``` 2. If you can't find the `.aws/config` file, Run the following command entering appropriate information from above ``` aws configure --profile AnnotationContributor ``` After the configuration is set you should be able to upload resources using ``` # To upload a full directory use recursive: aws --profile AnnotationContributor s3 cp test_dir s3://annotation-contributor/test_dir --recursive --acl public-read # To upload a single file aws --profile AnnotationContributor s3 cp test_file.txt s3://annotation-contributor/test_file.txt --acl public-read ``` Please upload the data with the appropriate directory structure, including subdirectories as necessary (i.e. top directory must be software package name, then if applicable, subdirectories of versions, ...). Please also do not forget to use the flag `--acl public-read`; This allows read access to the data file. Once the upload is complete, email hubs@bioconductor.org to continue the process. To add the data officially the data will need to be uploaded and the metadata.csv file will need to be created in the github repository. # Validating The best way to validate record metadata is to read inst/extdata/metadata.csv with `ExperimentHubData::makeExperimentHubMetadata()`. If that is successful the metadata are ready to go. # Example metadata.csv file and more information As described above the metadata.csv file (or multiple metadata.csv files) will need to be created before the data can be added to the database. To ensure proper formatting one should run `AnnotationHubData::makeAnnotationHubMetadata` on the package with any/all metadata files, and address any ERRORs that occur. Each object uploaded to S3 should have an entry in the metadata file. Briefly, a description of the metadata columns required: * Title: 'character(1)'. Name of the resource. This can be the exact file name (if self-describing) or a more complete description. * Description: 'character(1)'. Brief description of the resource, similar to the 'Description' field in a package DESCRIPTION file. * BiocVersion: 'character(1)'. The first Bioconductor version the resource was made available for. Unless removed from the hub, the resource will be available for all versions greater than or equal to this field. This generally is the current devel version of Bioconductor. * Genome: 'character(1)'. Genome. Can be NA. * SourceType: 'character(1)'. Format of original data, e.g., FASTA, BAM, BigWig, etc. 'AnnotationHubData::getValidSourceTypes()' list currently acceptable values. If nothing seems appropiate for your data reach out to hubs@bioconductor.org * SourceUrl: 'character(1)'. Location of original data files. Multiple urls should be provided as a comma separated string. If the data is simulated we recommend putting either a lab url or the url of the Bioconductor package. * SourceVersion: 'character(1)'. Version of original data. * Species: 'character(1)'. Species. For help on valid species see 'getSpeciesList, validSpecies, or suggestSpecies.' Can be NA. * TaxonomyId: 'character(1)'. Taxonomy ID. There are checks for valid taxonomyId given the Species which produce warnings. See GenomeInfoDb::loadTaxonomyDb() for full validation table. Can be NA. * Coordinate_1_based: 'logical'. TRUE if data are 1-based. Can be NA. * DataProvider: 'character(1)'. Name of company or institution that supplied the original (raw) data. * Maintainer: 'character(1)'. Maintainer name and email in the following format: Maintainer Name . * RDataClass: 'character(1)'. R / Bioconductor class the data are stored in, e.g., GRanges, SummarizedExperiment, ExpressionSet etc. * DispatchClass: 'character(1)'. Determines how data are loaded into R. The value for this field should be 'Rda' if the data were serialized with 'save()' and 'Rds' if serialized with 'saveRDS'. The filename should have the appropriate 'rda' or 'rds' extension. The function 'AnnotationHub::DispatchClassList()' will output a matrix of currently implemented DispatchClass and brief description of utility. If a predefine class does not seem appropriate contact hubs@bioconductor.org * Location_Prefix: 'character(1)'. Do not include this field if data are stored in the Bioconductor AWS S3; it will be generated automatically. If data will be accessed from a location other than AWS S3 this field should be the base url. * RDataPath: 'character(1)'.This field should be the remainder of the path to the resource. The 'Location_Prefix' will be prepended to 'RDataPath' for the full path to the resource. If the resource is stored in Bioconductor's AWS S3 buckets, it should start with the name of the package associated with the metadata and should not start with a leading slash. It should include the resource file name. * Tags: 'character() vector'. 'Tags' are search terms used to define a subset of resources in a 'Hub' object, e.g, in a call to 'query'. Any additional columns in the metadata.csv file will be ignored but could be included for internal reference. More on Location_Prefix and RDataPath. These two fields make up the complete file path url for downloading the data file. If using the Bioconductor AWS S3 bucket the Location_Prefix should not be included in the metadata file[s] as this field will be populated automatically. The RDataPath will be the directory structure you uploaded to S3. If you uploaded a directory `MyAnnotation/`, and that directory had a subdirectory `v1/` that contained two files `counts.rds` and `coldata.rds`, your metadata file will contain two rows and the RDataPaths would be `MyAnnotation/v1/counts.rds` and `MyAnnotation/v1/coldata.rds`. If you host your data on a publicly accessible site you must include a base url as the `Location_Prefix`. If your data file was at `ftp://myinstiututeserver/biostats/project2/counts.rds`, your metadata file will have one row and the `Location_Prefix` would be `ftp://myinstiututeserver/` and the `RDataPath` would be `biostats/project2/counts.rds`. This is a dummy example but hopefully it will give you an idea of the format. Let's say I have a package myExperimentPackage and I upload two files one a SummarizedExperiments of expression data saved as a .rda and the other a sqlite database both considered simulated data. You would want the following saved as a csv (comma seperated output) but for easier view we show in a table: Title | Description | BiocVersion | Genome | SourceType | SourceUrl | SourceVersion | Species | TaxonomyId | Coordinate_1_based | DataProvider | Maintainer | RDataClass | DispatchClass | RDataPath ---------------|----------------------------------------------|-----|-------------|-----|----------------------------------------------------------------------------------------------------|------------|-------------|------|------|---------|-------------------------------------------------------|-----------|----------|-------------------------------------------------------------------------- Simulated Expression Data | Simulated Expression values for 12 samples and 12000 probles | 3.9 | NA | Simulated | http://mylabshomepage | v1 | NA | NA | NA | http://bioconductor.org/packages/myExperimentPackage | Bioconductor Maintainer | SummarizedExperiment | Rda | myExperimentPackage/SEobject.rda Simulated Database | Simulated Database containing gene mappings | 3.9 | hg19 | Simulated | http://bioconductor.org/packages/myExperimentPackage | v2 | Home sapiens | 9606 | NA | http://bioconductor.org/packages/myExperimentPackage | Bioconductor Maintainer | SQLiteConnection | SQLiteFile | myExperimentPackage/mydatabase.sqlite