Scroll to navigation



load2db - load genome sequences and extrinsic evidence hints into a MySQL database


load2db [OPTIONS] --species=SPECIES --dbaccess=dbname,host,user,passwd inputfilename


load2db is a tool to load genome sequences and extrinsic evidence hints into a MySQL database.
This database must be created before calling load2db.
When storing genomes/hints of multiple organisms call this program repeatedly for each one.


inputfilename refers to a genome file in FASTA format or a hints file in GFF format.

Mandatory options

-s, --species=SPECIES

SPECIES is a name for your species and the same identifier as is used in the treefile and alnfile parameters to augustus.

-d, --dbaccess=dbname,host,user,passwd

dbname,host,user,passwd are the name of the SQL database, the host name or IP, the database user and password

Additional options

-c, --chunksize=size (⇐ 1000000)

This option is only relevant when loading a sequence file.
The sequences in the input genome are split into chunks of this size so that subsequent retrievals of small sequence ranges do not require to read the complete - potentially much longer - chromosome. Default Value: 50000

-h, --help

Produce help message.


create database before calling load2db:
$ mysql -u root -p
> CREATE DATABASE vertebrates;
> CREATE USER 'myuser'@'%' IDENTIFIED BY 'mypasswd'; /* or whatever the password code is*/
> GRANT ALL ON vertebrates.* TO 'myuser'@'%' IDENTIFIED BY 'mypasswd';
load2db --species=mouse --dbaccess=vertebrates,localhost,myuser,mypasswd mouse.fa
load2db --species=mouse --dbaccess=vertebrates,localhost,myuser,mypasswd mouse.hints.gff
load2db --species=human --dbaccess=vertebrates,localhost,myuser,mypasswd human.fa
load2db --species=human --dbaccess=vertebrates,localhost,myuser,mypasswd human.hints.gff


AUGUSTUS was written by M. Stanke, O. Keller, S. König, L. Gerischer and L. Romoth.