#ARDock server implementation
git clone https://github.com/glaunay/ardock.git
npm install
Check the config file
##Tests ####1. Basic pdb parsing Test with the default input PDB file (./scripts/test/4MOW.pdb)
$node index.js --conf scripts/package.json --pdb --test
Specify a valid pdb file
$node index.js --conf scripts/package.json --pdb -f scripts/test/1BRS.pdb
####2. Computing engine The slurm and local implementations require netcat (nc command) to be installed
#####2.1 Local machine (multi-threaded) implementation
$node index.js --conf scripts/package.json --pdb -p 3 --local --test --slurm
Use the --test
flag to mimic a docking computation on the pdb file scripts/test/4MOW_D.pdb
#####2.2 Cluster of machines implementation (SLURM only) Perform a 10 probes surface sampling of the 1BRS.pdb A,B dimer
$node index.js --conf ./scripts/package.json --pdb -f ./scripts/test/1BRS.pdb -c A,B --slurm -p 10
####3. Running the HTTP server
#####3.1 Start the HTTP server on test mode
$node index.js --conf ./scripts/package.json --test --http
See result at http://ardock.ibcp.fr/test Node server listens on port 3000
#####3.2 HTTP Server with supports for Interactive web-client and REST interfaces
node index.js --conf ./scripts/package.json --http --slurm --rest -p 3
######3.2.1 REST API specifications Using the REST interface to trigger the download of a structure and process it. Results will be displayed as an annotated protein structure file
eg: http://ardock.ibcp.fr/rest?pdbCode=1BRS&chains=A,B
- pdbCode is valid PDB identifier (mandatory)
- chains is a list of chain identifiers separated by coma (optional, by default all valid chains are processed)
######3.2.2 Web-client specifications
IO Duplex connection communication protocol
Usage
Build the client code
$browserify -t browserify-css app.js -o js/bundleTest.js