To get the track hub address from the above link:

• Click on an item in column labeled "Hub Name"
• Copy the URL shown in the browser

This URL is the "track hub URL".

Philip found this API for retrieving the public hubs: https://api.genome.ucsc.edu/list/publicHubs

Please see my current implementation deployed here. Any feedback or suggestions are welcome.

The respective code can be found here.

Redeploying branch IGBF-2898 from repository https://bitbucket.org/pbadzuh/bioviz_pbdev.git to https://bioviztest.bioviz.org using new playbook developed in IGBF-2904 "Configure playbooks to specify deployed facade site for bioviz".

Using inventory file:

[targets]
bioviztest ansible_host=52.4.208.199 domain=bioviztest.bioviz.org stack=Develop  igb_branch=release-9.1.8 igb_version=9.1.8 igb_repository=https://bitbucket.org/lorainelab/integrated-genome-browser bioviz_repo=https://bitbucket.org/pbadzuh/bioviz_pbdev.git bioviz_version=IGBF-2898

[targets:vars]
ansible_ssh_common_args="-o StrictHostKeyChecking=no"
ansible_python_interpreter=/usr/bin/python
ansible_ssh_user=ec2-user

See attached edit screen capture image (file name 2021-08-10-Suggestions.png) for suggested improvements and possible bug.

Update: The "copy" and "refresh" icons did not work for me at first, but then I did a "hard reload" of the page, and now they do work. So I'm now able to use the new version of the page to generate new Quickload data sources.

Additional feature requests:

• Can you allow searching by genome version name? If I am looking at my own data and would like to compare it to data sets from other groups, it would be helpful to be able to search using the name of the reference genome I'm working with, e.g., "hg38", the synonym for IGB genome version name H_sapiens_Dec_2013.

• Can you modify the interface so that the section of the page titled "Next, add the URL below as a new data source in IGB" always appears immediately below the "Convert" button?

• Add a parenthetical after each species name indicating the genome version (or versions) available for that species in each Track Hub. For example: "Home sapiens (hg38)"

• Just a reminder: the example link needs to be updated. Also, after the example link, we should probably let people know which genome they should open in IGB. The example used currently refers to the older version of the human genome. Maybe we ought to pick a more up-to-date one, something that would be interesting enough to serve as a good demonstration.

Just wanted to make a comment: The table is really great. It makes the software 10 times more useful!

Currently I am using the new table to look for RNA-Seq or transcriptome data sets I can use to investigate whether or not the homeobox-minus MEOX1 splice variant, which is lowly expressed in healthy human tissues, is more highly expressed in cancerous tissues, a follow-up from the use case scenario in this IGB slidedeck.

Request:

• Prepend the protocol to the copied text so that users can copy and paste it into the IGB window without having to type of protocol string manually.

For today's walkthru, deployed latest IGBF-2898 branch to bioviztest.bioviz.org and latest main branch to TestHub1.bioviz.org.

attn: Philip Badzuh

Error:

When I click the "copy" arrow under the text entry box to try out the example track hub, the interface reports and error: "Please provide a valid track hub URL. See the example below for more details."

Two problems:

• The example does not work
• The error message says "See the example below for more details" but there are no details or help provided.

Testing branch IGBF-2898 from https://bitbucket.org/pbadzuh/bioviz_pbdev.git, commit:

decdf61 (HEAD -> IGBF-2898, origin/IGBF-2898) IGBF-2892: Cache public hub data beyond single user session

The table reports "Unknown" followed by genome name in every cell, in the column "Organisms and Genomes".

Do this for the next iteration:

• Change the column heading to "Genome Versions"
• Do not report "Unknown" for anything. If there is nothing to report, print nothing.
• If you can identify the organism or a given genome, print its species name in parentheses, e.g., hg19 (Homo sapiens). If not, print the name of the genome version.

Usability improvements - let's try these:

• Add a column that numbers the rows - this should be column 1. Scientists (our users) are quantitatively minded. They will want to be able to know how many track hubs contain data for the genome version they are using in their research, or the number of rows returned from a search. This will also help us test the interface for correctness.

• Move "actions" column to the front of the table, after the numbering column. As users learn the interface and the available Track Hubs, they will barely look at the rest of the information.

Merging existing changes to separate the below tasks into individual tickets.

TODO:

• fix url validation
• open links in new tab
• make tooltip and heading text more clear
• change genome column's name and text content
• make tooltip duration longer
• move actions to table left - add fill arrow icon
• move result section above table
• ensure result section is in view during/after conversion
• add info to result URL section
• number rows
• prevent result URL wrapping on hyphen

Backend:

• modify synonyms endpoint to ensure proper genome name in IGB header/topbar
• add hub validation endpoint

Please see PR here.

Thank you Philip Badzuh for the TO-DO and Backend lists in the next-to-last comment before this one. Next step is to make individual JIRA issues for each one, using the bullet point text as titles, with maybe one or two minor edits, e.g., "Number the rows"

Merged and deployed for testing.

Tested that the code properly deployed by reloading the page and observed that the "Track hub" link in the upper menu disappeared. Moving to closed.