Last modified 16 months ago Last modified on 14/10/10 11:47:50

  1. 1 Get the Ensembl database (Optional)
    1. 1.1 Download the Ensembl MySQL database
    2. 1.2 Create the database
    3. 1.3 Load the database structure
    4. 1.4 Load the data
  2. 2 Install Perl modules
  3. 3 Update properties files
    1. 3.1 Update <MINE>.properties
    2. 3.2 Add Ensembl to the list of datasources to be integrated.
  4. 4 Load data
    1. 4.1 Generate XML file
    2. 4.2 Load XML file into database

This page describes how to load Ensembl into your InterMine-bio database.

1 Get the Ensembl database (Optional)

First you will need the data from Ensembl, which are available via MySQL databases. Ensembl has a publicly available  MySQL database you can use. If you think you are going to be retrieving a lot of data from Ensembl or reliability is very important, it will likely be in your best interest to have a local database.

The following are instructions on how to load a local copy of an Ensembl database. You must have MySQL installed and correctly configured.

1.1 Download the Ensembl MySQL database

  •  ftp://ftp.ensembl.org/pub/current_mysql/
  • Or you can use our Perl script
    • We use get_ensembl_mysql to download and unzip data
    • To use this script, you need to install the appropriate Perl modules. See InterMinePerl.
    • The script requires three parameters - download_directory organism_name which_database eg: 
      # in bio/scripts
./get_ensembl_mysql /MY_DATA_DIR/ensembl homo_sapiens core

1.2 Create the database

# in mysql
create database homo_sapiens_core_59_37d;

1.3 Load the database structure

mysql -h HOST -u USERNAME -p homo_sapiens_core_59_37d < /MY_DATA_DIR/ensembl/homo_sapiens/homo_sapiens_core_59_37d/homo_sapiens_core_59_37d.sql

1.4 Load the data

Run this command in the same directory as the data you just downloaded: 

mysqlimport -h HOST -u USERNAME -p homo_sapiens_core_59_37d -L *.txt

See also:  MySQL

2 Install Perl modules

InterMine's Ensembl converter uses Ensembl's Perl API. Follow Ensembl's instructions for how to install the necessary Perl modules:

You will also need to install InterMine's Perl modules. Follow these instructions:

3 Update properties files

3.1 Update <MINE>.properties

You'll need one entry for every organism. The perl script run in Step 4.1 uses these entries to ascertain the location of the databases. For example: 

# core database
db.ensembl.9606.core.datasource.serverName=SERVER_NAME
db.ensembl.9606.core.datasource.databaseName=homo_sapiens_core_59_37d
db.ensembl.9606.core.datasource.species=homo_sapiens
db.ensembl.9606.core.datasource.user=DB_USER
db.ensembl.9606.core.datasource.password=DB_PASSWORD

# variation database
db.ensembl.9606.variation.datasource.serverName=SERVER_NAME
db.ensembl.9606.variation.datasource.databaseName=homo_sapiens_variation_59_37d
db.ensembl.9606.variation.datasource.species=homo_sapiens
db.ensembl.9606.variation.datasource.user=DB_USER
db.ensembl.9606.variation.datasource.password=DB_PASSWORD

3.2 Add Ensembl to the list of datasources to be integrated.

This is located in the project.xml file, and it should look something like: 

    <source name="ensembl" type="ensembl"> 
     <property name="src.data.dir" location="/MY_DATA_DIR/ensembl"/> 
    </source>

When you run a database build, every XML file in this directory will be loaded into the database. Currently FlyMine loads Ensembl data for Anopheles gambiae. See FlyMine's project.xml

4 Load data

4.1 Generate XML file

  • Run this command in /bio/scripts 
    ./ensemblAPI.pl MINE_NAME TAXONID /MY_DATA_DIR/ensembl
    for example: 
    ./ensemblAPI.pl flymine 7165 /data/ensembl/current

4.2 Load XML file into database

  • Run a build. The entry in the project.xml will instruct the build process to load the XML files you created in Step 1 into the database. For example, run this command in MINE_NAME/integrate: 
    ant -v -Dsource=ensembl