Installed the new Early Access release and tested on my Mac, the import button is working as expected. The reads are colored after a few seconds and the text file import works whether the header is present or not.

Closing this ticket!

PR is merged. New installers are built and deployed to bioviz.org Early Access section.

Ready for testing.

PR Submitted. https://bitbucket.org/lorainelab/integrated-genome-browser/pull-requests/1074

This is much clearer now, thank you Karthik! Recommending PR!

Changes above have been implemented. See commit

The Edit Tag and Color table starts with a 100 rows. When a file is imported, the table adds rows with respect to rows in the file. So the additional rows that was observed is the initial 100 rows. However, the light grey color pre-selected can be confusing. Will fix that in next commit.

Tested Karthik's branch on Mac by following the above protocol. The Color by table populates with the values from the file as expected. However, numerous (>25) rows were also appended to the end of the table after importing this file even though there's only one blank line in the file. These rows have an empty Tag Value column and Color set to grey.

Added an import button to Color by SAMtags window. See commit

Testing Protocol:

1. Load H_sapiens_Feb_2009 in Human genome (hg19)
2. Load 3k PMBC from a Healthy Donor.zip to Data Sources (see attachments to ticket)
3. Go to gene GNLY using the search function in top left
4. Right click in the track title -> Choose Color by -> Choose SAMtags
5. Select CB as tag value and click Import button
6. Select the [^pbmc_cel_color_info.txt] from the File Chooser and click ok
   # Expected behavior: Edit Tag and Color table pops up with all the values from the file
7. Click on Save and Apply and Click OK
8. Wait until the reads are colored.

A version of the import button was demo to Dr. Loraine and Dr. Freese. Takeaways and observations from that meeting:

1) When reads don't match the tag values provided by the user, its currently colored light grey. However, we found reads that may not have the tag itself . There was a suggestion that reads like that should be light grey and the other unmatched reads should be regular grey. However, if users are creating the import text file from a R program they can filter out reads that does not have the tag at all in the metadata.
2) The Edit Tags and Colors table is not scrollable right now. Needs to be fixed
3) The buttons and table header in Edit Tags and Colors window changes sizes when the window is resized. Needs to be addressed.
4) One of the conditions for the import button workflow is checking the headers in the file. The file headers is user defined so cannot really predict that. We just need to check if there is a tag column and a color column in the file and map that to reads.