GenomeVIP is a web application platform for performing variant discovery on Amazon's Web Service (AWS) cloud or on local high-performance computing clusters.
- v1.0 (3a1e935)
- v0.1 (ac89e21) | initial release
Below are potential installation scenarios for GenomeVIP.
AWS prerequisites: Amazon Web Services account; input genomics data stored in EBS (or S3, such as the public 1000 Genomes Project).
-
AWS cloud only: a pre-configured GenomeVIP server image is available in AWS public image repository. Computations created through this image will instantiate a public GenomeVIP runtime image with preinstalled genomics and supporting tools.
-
Local with AWS option: the GenomeVIP code is installed on a web server in the user's local network and configured to use locally installed genomics and supporting tools; computations can be sent to a local job manager (LSF support provided currently). This server may also be used as an alternative to the public server image mentioned in (1) above; AWS computations proceed as in (1) above.
A step-by-step user guide is in development and can be found in the document HowTo_GenomeVIP.docx
Login to your AWS account
Choose EC2 from the AWS Console
Select Images (AMIs)
From the dropdown public AMIs search GenomeVIP
Choose a server & runtime images
Launch the server (most recent version is recommended) from the Actions menu
Select Launch
Choose an instance type (m1.medium recommended)
Click "Next: Configure Instance Details"
Select number of instances (at least 1)
Select the network type, EC2 Classic
NOTE: The Availability Zone should match the instance zone (and the Volumes Availability Zone should match)
Skip the Storage & Tag tabs
Go to tab 6, Configure Security Group
Add a rule for HTTPS
Review selections, and launch
Choose the desired keypair (you can proceed without a key pair)
Go to your dashboard (orange box)
Select EC2 to check its running as many instances as you chose above and *get the public IP*
To get to the GenomeVIP homepage, the site is https://publicip/~genomevip (where publicip is the public IP of a running instance, the site is case sensititve).
Select either AWS or Local Cluster
If AWS
Provide your AWS login credentials
A session ID will be generated for future logins
If Local Cluster
Provide your login credentials to your local cluster
Select either an EBS volume (only with AWS accounts), user defined, or 1000 Genomes
Click “Load list” to load the data
Choose the .bam files to use
Choose a reference genome
Select if you want a copy of all sorted .bam and .bai created to storage directory
Execution Profiles
New
Select a Run mode (germline, somatic, de novo/ family trio)
Select Paramaters (these can be tuned further in each tool)
Existing
Upload from file
Select Genomic Regions
No selection
Standard chromosome1-22, X, Y
Standard chromosomes plus contigs MT, GI, GL, NC
User defined either manual entry or file upload
Click the “Apply Profile” button to apply above entries
Customization
For each tool switched on by the run mode selected, fine tune any parameters. If necessary, turn off any tools switched on or turn on any tools not automatically selected.
If AWS
Select an AWS computing resource instance
Select a storage bucket to put results
If Local Cluster
Select a local computing resource
Choose a working directory to run jobs
Select submit action (create script vs. execute script)
Specify any additional comments (optional)
Double Check
Click the “Preview” button to check execution profile
Click the “Validate” button to preview and check user settings
Click “Submit”
Check out the results by navigating to the location displayed (for both AWS and Local Cluster)
Enjoy!
Additional options and accounts management may be configured for AWS