BioGateway RDF generator

This is a code base implemented in Golang (making use of a couple of ancillary Python scrypts) for converting arbitrary datasets into RDF. It will download data from a specified set of data sources and generate RDF N-Triples files using a unified data model.

Installation

Clone the repository with:

git clone [email protected]:vlmir/bgw3.git

cd bgw3/src/adapters/

From within each subdirectory run:

go build

Move all the generated binaries to an external directory, e.g. ~/bin

Usage

mkdir myProject DAT OUT

**Note: myProject DAT OUT can be ANY directories outside bgw3

**Note: DAT & OUT will hold downloaded files & generated RDFs respectively

cd myProject

**Note: myProject will contain all STDOUTs and STDERRs

cp bgw3/misc/runAdaptors.sh ./

**Note: if you need to make changes to runAdaptors.sh it is more practical to do it here

Run the following command in the current directory to build a new set of RDF files

./runAdaptors.sh path_to_binaries path_for_downloading_files path_for_RDFs path_to_a_file_mapping_taxa_and_proteomes path_to_scripts OMIM_version

Example

./runAdaptors.sh ~/bin/ DAT/ OUT/ ~/repos/bgw3/misc/prm_txn.txt ~/repos/bgw3/misc/ 25 &

Output

subdirectories will be created in DAT for each data source filled with downloaded files

subdirectories will be created in OUT for each RDF graph filled with N-Triples files

The files are ready to be loaded into any implementation of RDF/triple store

The file bgw3/misc/loadBGW.sh can be used for uploading into a Virtuoso Open Source database