To get the available datasets for a genome use https://api.genome.ucsc.edu/list/tracks?genome=hg38 this API and implement a method similar to getAvailableDataSets(). Creating module classes from the JSON Response of the API.

Checking the get track data endpoint to investigate how to create the parser as it returns different JSON formats for different file types. Also, while checking this endpoint https://api.genome.ucsc.edu/list/tracks?genome=ailMel1 which returns available tracks/datasets for selected genome, found that there are some child nodes for a few datasets, which can be removed from the response by using this parameter trackLeavesOnly=1, also few dataset names have different genome name specified in it, need to connect with Nowlan Freese on this and clarify my doubts on how to parse this API.

As discussed with Nowlan Freese, will be working on only the genePred (refGene) file type first, after testing it will be moving to the other file types. Currently, working on fixing an error that I got while testing the getAvailableDatasets API implementation. Below is the error log:

10:12:15.661 [main] ERROR com.affymetrix.main.OSGiHandler - Unable to resolve ucsc-rest-api-service [82](R 82.0): missing requirement [ucsc-rest-api-service [82](R 82.0)] osgi.wiring.package; (osgi.wiring.package=org.apache.http.client.utils) Unresolved requirements: [[ucsc-rest-api-service [82](R 82.0)] osgi.wiring.package; (osgi.wiring.package=org.apache.http.client.utils)]
org.osgi.framework.BundleException: Unable to resolve ucsc-rest-api-service [82](R 82.0): missing requirement [ucsc-rest-api-service [82](R 82.0)] osgi.wiring.package; (osgi.wiring.package=org.apache.http.client.utils) Unresolved requirements: [[ucsc-rest-api-service [82](R 82.0)] osgi.wiring.package; (osgi.wiring.package=org.apache.http.client.utils)]

Added the dependencies in the shared-lib-wrapper which resolved the issue and resolved another issue by connecting with Dr.Freese. Working on the JSON mapping issue right now.

The issue I faced while doing the JSON mapping is that the JSON returned from the API https://api.genome.ucsc.edu/list/tracks?genome=hg38;trackLeavesOnly=1 has a dynamic key for the genome name which has the tracks (database info) as a sub-element in it. As the key name changes dynamically it's not possible to directly map the JSON object to a Java class, so I have used a setter method in the class to parse the JSON for the tracks alone separately by getting the JSON object using the genomeVersionName and initialize the tracks. Also, added a filter here in the tracks to only include genePred datasets/tracks as currently we will be implementing that alone and the API returns a lot of tracks with various types.

Tested the API implementation, working as expected. Available tracks (or datasets) are being loaded on application start. Code is available on the branch https://bitbucket.org/jaya-sravani/integrated-genome-browser/branch/IGBF-3501.

To test:

1. Download https://bitbucket.org/jaya-sravani/integrated-genome-browser/downloads/ installer or clone this branch to local https://bitbucket.org/jaya-sravani/integrated-genome-browser/branch/IGBF-3501 and start the application.
2. Select a particular genome.
3. See whether UCSC (Rest) datasets are visible in the Available data panel.

Testing on Mac:

1. Cloned Sravani's IGBF-3501 branch and started IGB.
2. Selected the galGal6 genome.
3. The UCSC (UCSC REST) datasets are visible in the Available Data window (augustusGene, ncbiRefSeqPredicted, ncbiRefSeqCurated, ncbiRefSeq)
4. There are no errors in the IGB logs.
   Note -> when I click Load Data no data are loaded. I assume this is because the data loading part has not been wired/implemented yet.

As discussed, closing this ticket because the data loading part is not implemented yet, will be done as part of this https://jira.bioviz.org/browse/IGBF-3528 subtask.