Participant selected topics

[martenson]

A list of more obscure topics participants are interested in

• [x] galaxy development (Moved to separate training.)
• [ ] launching a community
• [ ] IRIDA platform from Canada as an example of using Galaxy's backend without the interface.
• [ ] Migrating an instance on a stand alone server onto a HPC -- besides changing the config files to setup the runners, what other considerations should be taken
• [x] Interactive Environments: Jupyter notebooks (gxit)
• [ ] Getting data out of galaxy
• [ ] Archiving analyses (e.g. when paper is published, etc.)
• [ ] Collections
• [ ] Managing/Organizing the tool menu
• [x] Writing dynamic job's destinations (compute cluster)
• [ ] Overview of job working directory content (galaxy script, job script, metadata...)
• [x] Galaxy update best practices, including stopping/pausing jobs, etc.
• [ ] Debugging workflows, tools for users, including configuring galaxy to enable easier debugging, eg keeping job output dirs. Debugging tool output names in workflows.
• [x] Managing Data libraries
• [ ] Writing workflows: best practices.
• [ ] Including naming outputs, workflow version control (within Galaxy? git?), user and developer documentation for workflows (they are really pipelines, and need docs). Handling required parameters for tools in a workflow (marking them in UI? failing early?)
• [x] Using the API (bioblend)
• [x] sharing user contributed admin related scripts
• [ ] Testing tools and workflows: setting up continuous integration; common set of small, medium and large test input files that can be used across tools and workflows.

Already discussed

• tool HTML whitelist
• External authentication
• Active Directory authentication -- also how this affects ownership and sharing datasets.
• Reports webapp
• Conda environments for installing Galaxy, for running jobs and handle tools dependencies
• Backup best practices

[Slugger70]

We should probably mention the whitelist to the intro participants too.. It's simple and won't take long.. Maybe in a section where we mention the visualisations?

[clijeron]

Active Directory authentication -- also how this affects ownership and sharing datasets.

Migrating an instance on a stand alone server onto a HPC -- besides changing the config files to setup the runners, what other considerations should be taken?

Backup best practices?

[martenson]

@clijeron I added your items to the top level post, thank you!

[pdblood]

Interactive Environments: Jupyter notebooks

[lparsons]

• Getting data out of galaxy
• Archiving analyses (e.g. when paper is published, etc.)
• Collections
• External authentication
• Managing/Organizing the tool menu

[laure182]

• Writing dynamic job's destinations
• Conda environments for installing Galaxy, for running jobs and handle tools dependencies
• Overview of job working directory content (galaxy script, job script, metadata...)

[tshtatland]

Galaxy update best practices, including stopping/pausing jobs, etc.

[tshtatland]

Debugging workflows, tools for users, including configuring galaxy to enable easier debugging, eg keeping job output dirs. Debugging tool output names in workflows.

[laure182]

Managing Data libraries

[tshtatland]

Writing workflows: best practices. Including naming outputs, workflow version control (within Galaxy? git?), user and developer documentation for workflows (they are really pipelines, and need docs). Handling required parameters for tools in a workflow (marking them in UI? failing early?)

[lparsons]

Using the API and sharing user contributed admin related scripts

[tshtatland]

Testing tools and workflows: setting up continuous integration; common set of small, medium and large test input files that can be used across tools and workflows.

[martenson]

I updated the top level post with all comments.

[blankenberg]

building paired collections?

[joe-of-all-trades]

More explanation on how pulsar works. For example, I want to offload TOPHAT2 operation from the Galaxy server to a remote Pulsar server. How do I tell Pulsar to run TOPHAT2 on the dataset against a specific genome using multiple cores? Do I simply need to configure this on the galaxy server and the dependencies ( TOPHAT2, and genome ) will be fetched automatically on the Pulsar server?

In today's exercise, was a copy of custom tool saved into the pulsar folder ?

[martenson]

Results of strawpoll.