This devlog covers the final preparation work for the first JU SeqWorkbench Alpha package and the development priorities expected immediately after release.
The current application version is 0.1.0-alpha.
This page is still a development note rather than a sales or release announcement. The first public alpha package has not yet been posted.
Recent release-preparation work has focused less on adding new analysis features and more on making the existing workflow safer and easier to test as a packaged desktop application.
Current preparation includes:
The current source-level alpha smoke suite passes 29 scripts/smoke_*.py checks in the latest release-preparation run.
JU SeqWorkbench is being prepared as a Windows desktop application using PySide6 and PyInstaller.
The packaging goal is to keep the application core and external analysis executables clearly separated.
For the current alpha workflow:
This separation also makes it easier to keep external-tool troubleshooting and license obligations distinct from the application core.
Open-source license review remains part of release preparation.
The current design intentionally keeps external alignment programs outside the application package and communicates with them through subprocess and temporary FASTA input/output files.
The packaged application itself also needs appropriate third-party license/notice material for libraries that are redistributed with it.
The goal of the alpha is not to claim that license review is permanently finished. The goal is to document the current dependency boundary clearly and prepare a package that can be reviewed and updated before broader or commercial distribution.
JU SeqWorkbench is designed as a local desktop workflow.
Sequence files, working projects, analysis tables, and local external-tool execution are intended to remain on the user’s machine unless the user separately chooses to export or share data elsewhere.
This is particularly relevant for research workflows where unpublished or institution-restricted sequence data should not be uploaded to an external AI or web service simply to perform routine inspection.
The first alpha is intentionally narrower than the long-term plan.
Its purpose is to validate whether the existing connected workflow is useful:
FASTA / AB1
→ sequence review and editing
→ external MSA / alignment review
→ grouping / typing / similarity inspection
→ Point or Region Visualization
→ table / figure / FASTA export
The first alpha should therefore be evaluated mainly on:
Alpha feedback will explicitly ask for visualization and analysis ideas, not only errors.
Useful requests include:
Screenshots or references from papers and other tools can be useful design references, provided no confidential sequence data is shared publicly.
The first update cycle after the public alpha is expected to focus on three areas before expanding into a broader list of features.
The current Point Visualization workflow is strongest when users already know that aligned positions are biologically comparable.
More complex or highly divergent viral datasets may require an earlier relationship layer. The same named ORF or corresponding segment does not automatically guarantee that direct site-by-site comparison is the best first analysis.
The post-alpha plan therefore prioritizes exploration of:
The intended direction is closer to:
Sequence set
→ relationship / similarity inspection
→ select comparable clade / ORF / segment / region
→ alignment review
→ point or region analysis
→ visualization / export
These features are planned work and are not presented as implemented in 0.1.0-alpha.
The existing visualization layer is useful for alpha workflow validation, but visual presentation itself will be treated as a major development target.
Current AI-assisted plotting and analysis tools can generate polished figures quickly, which has raised user expectations for clarity and visual quality. The goal for JU SeqWorkbench is not to compete by generating decorative charts. It is to make scientific sequence views easier to read, compare, and connect back to the underlying data.
Planned investigation includes:
Richer analysis and visualization increase rendering cost.
Post-alpha work will therefore also review performance in the viewer, result tables, mutation maps, and other large visual outputs.
Planned areas include:
The goal is to improve visual quality without making the application feel heavier or less responsive.
Group-to-group comparison remains an important planned analysis feature and should connect naturally with the relationship layer.
For example, a future workflow may allow a user to identify two clades or user-defined groups and then compare site/region mutation patterns between them.
Other planned directions such as deeper annotation support, more external tools, and additional reporting features remain relevant, but they are not the first priority immediately after the initial alpha release.
The first alpha does not need to contain every planned feature.
The immediate goal is to publish a coherent build, collect structured feedback, and then update the program quickly around the most important missing analysis layer and visualization needs.
That is why sequence relationship / ORF / segment-aware analysis, visualization modernization, and rendering performance are being documented now as the first major post-alpha priorities rather than being rushed into the initial package.