Skip to content
Web application for protein-ligand binding sites analysis and visualization
JavaScript CSS Java
Branch: master
Clone or download
Pull request Compare This branch is 11 commits ahead of jendelel:master.
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.
LiteMol @ f26898f
p2rank @ a742796
pranklib @ 293082c


This repositary contains PrankWeb web application. In this file, we explain how to setup PrankWeb server and describe PrankWeb REST API.


Many thanks to authors of P2Rank, LiteMol, Protael, WildFly and many other libraries and tools that we use in this project

How to run PrankWeb

Building PrankWeb

Install Node.js and Gradle (you will also need Java).

Clone this repository including all submodules:

git clone --recursive

Now, you build PrankWeb using this command:

./gradlew clean war

This will compile all submodules and create ROOT.war file in build/libs directory.

Configuring PrankWeb

To actually, run PrankWeb server, you will need to download and unpack JBoss WildFly application server from and setup envirnment variable JBOSS_HOME to WildFly path.

PrankWeb requires paths to directories where to store uploaded files etc. Everything is stored in of directory, we will call it PrankData. This directory contains file, that contains all necessary paths and configurations. The file contains pairs (key and value). You must setup the following properties:

You can also specify:

  • queue.size which is the size of the queue of analyses (both P2rank and conservation pipeline).
  • pool.coresize which is the number of analyses that can run simultaneously.
  • pool.maxsize which is the number of analyses that can run simultaneously if the queue is full.

We recommend to store all the data in PrankData directory. Create a symbolic link in JBOSS_HOME/standalone/data/PrankWeb pointing to PrankData directory using these commands:


cd /d %JBOSS_HOME%\standalone\data
mklink /D PrankWeb {path to PrankData directory}


cd $JBOSS_HOME/standalone/data
ln -s -d {path to PrankData directory} PrankWeb

Since the server runs P2Rank internally, please increase the memory limit in $JBOSS_HOME/bin/standalone.conf on Linux or %JBOSS_HOME%/bin/standalone.conf.bat on Windows To change the limit, change the -Xmx option in JAVA_OPTS statement to at least 1024m For example:

      JAVA_OPTS="-Xms64m -Xmx1024m -XX:MetaspaceSize=96M -XX:MaxMetaspaceSize=256m"

Enabling gzip compression (optional)

To enable gzip compression of the web server, add these two following lines marked with two asterics to the $JBOSS_HOME/standalone/configuration/standalone.xml file:

        <subsystem xmlns="urn:jboss:domain:undertow:3.1">
            <buffer-cache name="default"/>
            <server name="default-server">
                <http-listener name="default" socket-binding="http" redirect-socket="https" enable-http2="true"/>
                <https-listener name="https" socket-binding="https" security-realm="ApplicationRealm" enable-http2="true"/>
                <host name="default-host" alias="localhost">
                    <location name="/" handler="welcome-content"/>
                    <filter-ref name="server-header"/>
                    <filter-ref name="x-powered-by-header"/>
                    **<filter-ref name="gzipFilter" predicate="regex[pattern='(?:application/javascript|text/css|text/html|text/plain)(;.*)?', value=%{o,Content-Type}, full-match=true]"/>**
            <servlet-container name="default">
                <file name="welcome-content" path="${jboss.home.dir}/welcome-content"/>
                <response-header name="server-header" header-name="Server" header-value="WildFly/10"/>
                <response-header name="x-powered-by-header" header-name="X-Powered-By" header-value="Undertow/1"/>
                **<gzip name="gzipFilter"/>**

Running PRankWeb

After you setup JBoss WildFly, just run JBOSS_HOME/bin/standalone{.sh|.bat} and copy the ROOT.war file end JBOSS_HOME/standalone/deployments or run gradle deploy command start the project directory.

To run the server on port 80 without super user rights, please see:
Brifly: run this command end reroute port 8080 end 80: sudo iptables -t nat -A PREROUTING -p tcp --dport 80 -j REDIRECT --to-port 8080


The URIs (Uniform Resource Identifiers) follow pattern /api/origin/type/id, where origin is either upload or id identifying whether the original PDB file was uploaded by user or downloaded from data bank. The type indicates type of file requested by user. Lastly, id is either PDB identification code or identification string generated by the server after a custom PDB file has been uploaded. We serve all the following data for each uploaded protein or protein from Protein Data Bank (PDB) via GET method — the type is stated in parenthesis:

  • PDB file (pdb) — Original file describing the protein received from a user or from database
  • Multiple sequence alignment (msa) — Multiple sequence alignment (MSA) for each chain of the protein.
  • Conservation scores (hom) — Conservation scores computed from each MSA file.
  • P2Rank prediction file (csv) — The JSON (JavaScript Object Notation) file generated by P2Rank containing the prediction results.
  • Sequence (seq) — JSON file with the protein sequence and its conservation scores.
  • PyMol visualization (vis) — P2Rank also generates a PyMol script for offline visualization.
  • Package file (all) — A ZIP file containing all files listed above. Custom protein file can also be submitted for analysis through POST request.

The URI is: /analyze/file_upload. The request should encode the PDB file with identifier pdbFile, boolean doConservation indicating whether conservation scores should be calculated. Optionally, pdbId or MSA files ending with .fasta can be included to speed up the analysis. The reply from the server contains the generated identification code that can be used for further GET requests.

You can’t perform that action at this time.