This is a service to assemble genomes and metagenomes with user's choice of assembly software. It currently supports more than 20 assemblers and processing modules.

REST API

• start with a fresh KBase image (last tested on v15) with security group 'default' or 'assembly-rast-group'
• log in as ubuntu and get root access with 'sudo su'
• enter the following commands:

cd /kb
git clone ssh://kbase@git.kbase.us/dev_container
cd /kb/dev_container
./bootstrap /kb/runtime
source user-env.sh
cd /kb/dev_container/modules
git clone ssh://kbase@git.kbase.us/assembly
git fetch origin
git checkout RC
make deploy
make test

• start with a fresh KBase image (last tested on v14) with security group 'assembly-rast-group'
• log in as ubuntu and get root access with 'sudo su'
• enter the following commands:

cd /kb
git clone ssh://kbase@git.kbase.us/dev_container
cd /kb/dev_container/modules
git clone ssh://kbase@git.kbase.us/assembly
cd /kb/dev_container
./bootstrap /kb/runtime
source user-env.sh
make deploy-service

The current deployment first invokes scripts/install_dependencies.sh. This will be no longer necessary once a kbase image is built with these dependencies.

• the assembly service includes one control server and at least one compute server
• for now, we assume a long-running compute server, and use 'service' to refer to the control server
• to start and stop the service, use the 'start_service' and 'stop_service' scripts in /kb/deployment/services/assemly
• on test machines, assembly service listen on port 5672, so this port must be open
• after starting the service, the process id of the service is stored in the 'service.pid' file in /kb/deployment/services/assembly/

• Client deployment
• Testing
• Support for more assemblers
• Support for large eukaryotic genome assembly
• Support for metagenome assembly on supercomputer