Cluster root nodes

[stefanhahmann]

Resolves #25

Already added:

• UI dialog for parameter selection
• Find all roots starting from the selected start timepoint
• Compute the similarity matrix
• Derive hierarchical clustering from the similarity matrix (average linkage)
• Derive the selected number of classes from the similarity matrix
• Create tagset from the computed classes.
• Assign a tag to each spot between start end endtimepoint
• Normalized and Average Zhang edit distance
• Unit tests
• Checkbox to select show dendrogram yes/no
• Handling of exceptions and proper feedback to ui
  • More classes than leaves
  • Negative start/end/number of classes/min cell divisions
  • Start >= End
• Parameter explanations in dialog
• Persist Input parameters
• Crop criterion start/end number of cells
• Show selected parameters in dendrogram title
• Ensure UI dialog is not opened again, when already open, but brought to front again
• Investigate small differences dendrograms (python/java) -> seems that java average linkage is giving correct results
• Adapt number of roots, if crop start > 0

[sonarcloud]

Kudos, SonarCloud Quality Gate passed!   

0 Bugs
0 Vulnerabilities
0 Security Hotspots
0 Code Smells

80.9% Coverage
0.0% Duplication

The version of Java (11.0.20) you have used to run this analysis is deprecated and we will stop accepting it soon. Please update to at least Java 17.
Read more here

[stefanhahmann]

The plugin works, but:

* The threshold line is drawn in the wrong place in the dendrogram view.

* The ClusterRootNodesView is not large enough when first shown such that the "create tag set) button is hidden. (This is a bug in scijava-ui-swing)

* The ClusterRootNodesView does not appear in the task bar. (Bug in scijava-ui-swing)

Thanks for the advices.

• I changed the place of the threshold.
• The button not being visible seems to be a bug of scijava-ui-swing in Unix based Systems. It does not occur on a Windows machine
• The task bar issue is not something that can be resolved in this issue.

For the sake of completeness I made a another comparison between the Python based results in the results in this Java implementation. It looks fairly similar now (apart from the order of the clusters, which differs, but is not really relevant):