Symap visualisation

Symap relies on MySQL, which can be downloaded from the MySQL site.

Installation of MySQL

MySQL required libaio in order to run, so install this first using APT.

 sudo apt-get install libaio1

Then install MySQL:

 sudo apt-get install mysql-server

This will bring up a setup screen:

Then there will be a stream of output to the terminal. Following this, create a database and create a user and password.

 sudo service mysql start
 mysql --user=root -p

This will prompt you for a password with:

  Enter password:

Use the password you set in the setup screen. Once in the MySQL shell, use the following commands to create a database. If you are the sole person using the database and in charge of the server, you can access it as root using the root password specified. Otherwise, there are instructions on the MySQL site to create users with different privileges and passwords. In this example, the database is called 'symap' as it will be used with the symap software.

 mysql>CREATE DATABASE symap

Running Symap

Symap requires a MySQL database for datasets larger than the example data, which comes with its own mini database. In order to access the symap database created in the previous step, open the symap.config file in the symap installation base directory.

 # SyMAP configuration parameters file
 #

 #########################
 # Required parameters:
 #########################

 # Database name, server, admin and public access info
 db_name             = symap
 db_server           = localhost
 db_adminuser        = root
 db_adminpasswd      = password
 db_clientuser       = root
 db_clientpasswd     = password

Note: 'password' is not a real password. This should be the password specified when setting up MySQL in the previous step.

In the install base directory, you can then run:

 ./symap

This will access the symap MySQL database.

If loading gets stuck on the annotations file, open the mysql configuration file:

 sudo vi /etc/mysql/my.cnf

Add the following lines to the end of this file:

 [mysqld]
 innodb_buffer_pool_size=1G
 innodb_flush_log_at_trx_commit=0

This will allow for sequences to be loaded, though it is still relatively slow. For a genome of 39 Mb with ~11000 annotations, loading took 1 hr 6 min. This was just for the initial load. Following this, sequences are stored in a mysql database and should load quickly. Note: When setting options in my.cnf, they should be preceded by [group id].