[Segway-announce] Segway 2.0.2 released
Eric.Roberts at uhnresearch.ca
Thu Mar 1 12:09:15 PST 2018
We are pleased to announce the availability of Segway 2.0.2!
There are a number of significant quality of life improvements and bug fixes in this minor release. This version is future-proofed against newer versions of numpy (>1.14). This release also fixes a case where window selection was incorrect in the case of using multiple genomedata archives. It is highly recommended to upgrade Segway to this release.
Segway now outputs it's job log (formally the jobs.tab file) into separate per-instance files. For example, the 1st training instance's job log will be now inside "jobs.0.tab". This allows tracking down problematic jobs much easier. Before all logs were in a single file and there was a potential race condition where each instance could potentially overwrite over previous logs.
A new file named "window.bed" is output with all Segway runs. This file contains a list of all regions (or windows) considered by Segway for the current analysis (training or identification). This can be cross referenced with the "jobs.instace.tab" to pinpoint exactly where in the genome the analysis is taking place per job. The "window.bed" from identify can be helpful as visual tool in your resulting annotations to see exactly where Segway starts and stops running its segmentation regardless of data presence.
Please let me know if you have any comments on Segway, its documentation, web site, installation, or anything else. (The preferred place to make these comments is on the segway-users mailing list or by reporting an issue on the issue tracker, both linked from the main web site).
Here is a full list of changes made since the last announcement here:
* segway: add window.bed output to cross reference window index to chromomsome
* segway: add instance specific job log files to avoid log overwrites
* segway: update support for versions of numpy > 1.14 by adding legacy scalar
and vector text printing
* segway: fix window selection when using multiple genomedata archives
* segway: fix regression of stdout not being captured from training jobs
Known Bugs: To see all known bugs head to the issue tracker at http://bitbucket.org/hoffmanlab/segway/issues
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Segway-announce