BIAFLOWS tutorial @ I2K 2020

Dear bioimage analyst / developer,

BIAFLOWS is a web platform that enables to encapsulate any bioimage analysis workflow so as to simplify its reproducible deployment and public benchmarking. BIAFLOWS-enabled workflows can run directly on our servers and the results can be explored and compared from a powerful multidimensional image viewer and interactive benchmark metrics results tables.

We believe that BIAFLOWS can greatly boost the visibility and promote the performance of your own workflows. We already support a wide range of bioimage analysis problems (e.g. nuclei segmentation, cell tracking, neuron tracing…) and we plan to extend this list and include new annotated datasets for benchmarking very soon.

If you are interested in knowing more and plan to include your own workflows to BIAFLOWS, you are invited to join the tutorial we organize twice at I2K 2020, respectively on 1st December 8:00 - 12:00 UTC and 2nd December 12:00:16:00 UTC.

You can find the program of the tutorial and apply here (deadline: November 20th):

and find more information on BIAFLOWS here:

https://biaflows.neubias.org

We hope to have you on board!

Best,

BIAFLOWS team

3 Likes

Hi all,

thanks so much for the very useful session! I reached point 8, and it seems that things work nicely, please see attached the screenshots from my dockerhub account. I would kindly ask you for some more details on pint 9 on, to complete the reaming steps of BIA workflow creation.

Best regards,
Ahmed

Hi @Ahmed_S_Khalil,

in point 8 you have configured dockerhub to ceate a new docker-image each time a new release of your workflow is created on github.

In point 9 you will create a new release of your workflow on github. This will automatically trigger the creation of the image on dockerhub and biaflows will automatically import the workflow from there.

To create a new release on github, go to your repository on github and look for the link “Releases” on the right side of the page (between “About” and “Packages”). Click on “Releases”, On the releases-page press the button “Draft a new release”. In the field “tag version” enter a new version number for the release (for example v1.0). On the bottom of the page press the button “Publish release” (title and description are optional).

In step 10 you just have to wait that dockerhub finishes to build the image for the new release. You can check the progress on the build-tab of project on dockerhub.

We will add the details and screenshots to the documentation and send a link here,
Best,
Volker

3 Likes

Thanks a lot @volker !