Skip to content

Latest commit

 

History

History
102 lines (76 loc) · 3.54 KB

README.md

File metadata and controls

102 lines (76 loc) · 3.54 KB
Raw

ModelSEEDDatabase in Solr

The primary purpose of this directory is to store and maintain the schema that we use for our ModelSEEDDatabase Solr cores.

The same files can be used in anyone's local instance of Solr, especially if they wish to maintain their own version of the database. So I explain here how to do this.

(The assumption is that the user is a sysadmin or has sudo access, I've not attempted to install/run solr without sudo access)

Installation of Solr

Follow the 'Solr Quick Start' link on the page:
http://lucene.apache.org/solr/guide/8_4/solr-tutorial.html

to Download/Unpack/Launch Solr on your Solr server machine, if locally, it will be running at: http://localhost:8983/solr/

As of 02/13/20, I got Solr 8.4.1 running using JDK 11 on Ubuntu 18.04 and found this guide helpful:
https://tecadmin.net/install-apache-solr-on-ubuntu/

As of 04/30/20, I was able to get Solr 8.5 working on my Macbook using MacPorts: https://www.macports.org/

For the purpose of this guide, I'm using the $SOLR_PATH environment variable, but it differed between the two environments. On Ubuntu, $SOLR_PATH is /opt/solr/ and on OS X, using MacPorts, $SOLR_PATH is /opt/local/share/java/solr-8.5.0/.

Create Solr Cores

Here we create the cores using the schema that is available in this directory

sudo -u solr ${SOLR_PATH}/bin/solr create_core -c compounds -d cores/compounds/
sudo -u solr ${SOLR_PATH}/bin/solr create_core -c reactions -d cores/reactions/

Load Biochemistry Data

Here we load the actual data:

${SOLR_PATH}/bin/post -c compounds ../Biochemistry/compounds.json
${SOLR_PATH}/bin/post -c compounds ../Biochemistry/reactions.json

Access the Data

You can query the data in a number of ways:

  1. Solr's default UI:
    http://localhost:8983/solr/#/compounds/query
    http://localhost:8983/solr/#/reactions/query

  2. ModelSEED-UI:
    The ModelSEED has a custom search interface for exploring the biochemistry:
    https://modelseed.org/biochem/reactions/

    The code for the web server is publicly available:
    https://github.com/ModelSEED/ModelSEED-UI

    and any user can use this code to host their own ModelSEED website. In doing so, they can point it to their local Solr instance, and use it to explore their biochemical data. To do so, you need to update a line in this file:
    https://github.com/ModelSEED/ModelSEED-UI/blob/master/config.js

    You have to change the variable solr_url so that its configured to point to
    http://localhost:8983/solr

  3. REST
    Apache Solr is RESTful, meaning it has an API that allows the server to be queried using http and any command-line tool or programming language. We have examples of this, in using curl and python in the examples folder:

examples/Check_Status_of_Cores.sh
examples/Check_Status_of_Cores.py
examples/Query_ModelSEED_Example.py
examples/Query_ModelSEED_Example.sh

KBase

The same biochemistry data hosted here and at https://modelseed.org is also available at KBase but the data is hosted by a different database. KBase has a python client that can be used to access the data, and the instructions for doing so are in the example code in the same folder:

examples/Query_KBase_Example.py

Authentication

We're using our own approach to restrict access to the Solr configuration, and so we are not using basic authentication, but I found this guide helpful for testing solr authentication:

https://docs.nginx.com/nginx/admin-guide/security-controls/configuring-http-basic-authentication/