We are pleased to announce a new release of the VDJServer website and associated tools.
This release was made available for public use on August 25, 2016.
- The pRESTO tool is now available. Much of the pRESTO functionality is available in a workflow supporting single-end and paired-end reads, length and quality filtering, barcoding, unique molecular identifiers (UMI), forward and reverse primers, and collapsing of unique sequences. VDJPipe is also automatically run during the worfklow to provide pre- and post-filtering statistics with corresponding chart visualizations.
- Execution levels are automatically determined for jobs. Previously jobs that ran on the Lonestar5 or Stampede supercomputer had their execution time set to 48 hours, which could cause the jobs to sit in the queue for awhile before being run. Now this time is optimized based upon the size of the input. Small and medium size jobs have more appropriate execution times defined, which allow most jobs to be run immediately. Users should notice jobs starting and finishing much quicker than before.
- VDJML Version 1.0.0 has been released and is integrated with VDJServer.
- Use paginated requests for retrieving data from Agave. For some projects with a large number of files (>100), sometimes not all files would appear. This has been resolved so now any number of files are supported.
- Resolved some bugs with the vdj-api server process.
- Fixed a duplicate file checking issue.
- We have written an extensive integration test suite that exercises all major functional and integration points in VDJServer. This will help increase the reliability of VDJServer.
- We have performed some reorganization of the Docker images for VDJServer. This facilitates running more development tests. We also upgraded to newer versions.