Note Oct 2018: This package has been deprecated and is no longer being actively maintained. Instead, users interested in accessing EuPathDB resources in R/Biocondcutor can now use the much simpler AnnotationHub interface to query for resources.

See the Accessing EuPathDB Resources using AnnotationHub vignette for examples.

This script uses resources from EuPathDB databases such as TriTrypDB and ToxoDB to generate a Bioconductor Organism annotation packages.

For this script to work, you must have installed the following R libraries:

• yaml
• rtracklayer
• GenomicFeatures
• OrganismDbi

The use this script, begin by modifying or creating a new YAML configuration file specifying the details for the species you wish to process. Examples configuration files for several species can be found in the settings folder. Next, modify the build_xx.r scripts to point to the setting file you wish you use.

Note that at present, you may also need to edit the build_orgdb.r script to enable it to properly map KEGG identifiers to EuPathDB identifiers for your target species. In the future, this will be generalized if possible so that this can be handled via config settings.

Now we are ready to construct the OrgDb and TranscriptDb packages:

$ Rscript build_orgdb.r
$ Rscript build_txdb.r
$ sh prepare_dbs.sh

This will generate OrgDb and TranscriptDb packages in the current working directory. The prepare_dbs.sh script performs some post-processing to replace default names with desired before generating the final Organism package.

The final step then is to construct an OrganismDb package. In order to proceed, however, you must first install the OrgDb and TranscriptDb packages generated above.

You can then use the install.packages command to install the annotations database locally, e.g.:

install.packages("./org.LmjF.tritryp.db", repos=NULL)

Finally, the Organism package can be constructed using the following command:

$ Rscript build_organismdb.r

All done!

This script has been designed to work with any EuPathDB organisms for which both GFF and gene TXT annotations are available. Currently, however, it has only been tested on L. major Friedlin annotations. If you encounter issues using it for your organism of interest, feel free to submit fixes via pull requests, or report the problem as an issue.

For questions, feel free to contact me at [email protected].

Below are a list of packages generated for serveral species using these scripts:

• TODO

• AnnotationForge - Making Organism packages
• Bioconductor - Making and Utilizing TxDb Objects
• Bioconductor - Making an OrganismDb package
• Bioconductor - AnnotationForge
• Bioconductor - GO.db
• Bioconductor - OrganismDbi