# Enhancement: STARsolo to support Velocyto (spliced/unspliced) matrices

**URL:** <https://help.galaxyproject.org/t/enhancement-starsolo-to-support-velocyto-spliced-unspliced-matrices/18035>\
**Category:** Uncategorized\
**Tags:** tool-dev\
**Created:** [June 10, 2026, 4:20pm UTC](https://help.galaxyproject.org/t/enhancement-starsolo-to-support-velocyto-spliced-unspliced-matrices/18035 "2026-06-10T16:20:19Z")\
**Posts on this page:** 3\
**Page:** 1

<div class="post-metadata">

**Author:** ![100517cqmu](https://sea2.discourse-cdn.com/flex020/user_avatar/help.galaxyproject.org/100517cqmu/32/8736_2.png) [@100517cqmu](https://help.galaxyproject.org/u/100517cqmu)\
**Post date:** [June 10, 2026, 4:20pm UTC](https://help.galaxyproject.org/t/enhancement-starsolo-to-support-velocyto-spliced-unspliced-matrices/18035/1 "2026-06-10T16:20:20Z")

</div>

Describe the bug  
When running the RNA STARsolo tool on [usegalaxy.org](http://usegalaxy.org) (tested on versions up to 2.7.11b+galaxy1 and 2.7.10b+galaxy4), the wrapper completely fails to export the Velocyto matrices (spliced.mtx, unspliced.mtx, ambiguous.mtx) to the user’s history, even when the Velocyto feature is explicitly selected.

To Reproduce  
Steps to reproduce the behavior:

Run RNA STARsolo with standard 10x scRNA-seq inputs (R1 & R2).

Under Collect UMI counts for these genomic features (–soloFeatures), select Velocyto. (Either alone or alongside Gene).

Run the tool.

Expected behavior  
The tool should output the standard Gene matrix PLUS the Velocyto matrices (preferably packaged as a dataset collection containing spliced/unspliced/ambiguous mtx files), exactly as the STAR C++ engine produces them.

Actual behavior (The Flaw)  
The job finishes successfully (turns green), but no Velocyto matrices appear in the history.  
By inspecting the Log.out (or tool stdout), we can confirm that the underlying STAR C++ engine correctly computed and saved the matrices in the temporary working directory (e.g., Solo.out/Velocyto/raw/).  
However, the Galaxy Tool XML wrapper lacks the necessary or tags in its section to discover and export these Velocyto files. The temporary directory is simply purged after the job finishes, destroying the highly valuable RNA velocity data.

Suggested Fix  
Please update the rna\_starsolo.xml wrapper. Add a dynamic output that triggers when ‘Velocyto’ in $solo.features is true, using \<discover\_datasets\> to harvest the spliced.mtx, unspliced.mtx, and ambiguous.mtx from the Solo.out/Velocyto/raw/ (or filtered) directory.

Thank you for maintaining these amazing tools! Fixing this will hugely benefit the spatial transcriptomics and scRNA-seq community relying on Galaxy for RNA Velocity analysis.

---

<div class="post-metadata">

**Author:** ![jennaj](https://sea2.discourse-cdn.com/flex020/user_avatar/help.galaxyproject.org/jennaj/32/27_2.png) [@jennaj](https://help.galaxyproject.org/u/jennaj)\
**Post date:** [June 12, 2026, 8:01pm UTC](https://help.galaxyproject.org/t/enhancement-starsolo-to-support-velocyto-spliced-unspliced-matrices/18035/2 "2026-06-12T20:01:04Z")

</div>

Welcome @100517cqmu

This sounds like a good enhancement request!

I see that you already made the ticket, so I’ll just link that back here → [CRITICAL BUG: STARsolo wrapper fails to output Velocyto (spliced/unspliced) matrices · Issue #8080 · galaxyproject/tools-iuc · GitHub](https://github.com/galaxyproject/tools-iuc/issues/8080)

Thanks! 🙂

---

<div class="post-metadata">

**Author:** ![wm75](https://sea2.discourse-cdn.com/flex020/user_avatar/help.galaxyproject.org/wm75/32/56_2.png) [@wm75](https://help.galaxyproject.org/u/wm75)\
**Post date:** [June 24, 2026, 9:55am UTC](https://help.galaxyproject.org/t/enhancement-starsolo-to-support-velocyto-spliced-unspliced-matrices/18035/3 "2026-06-24T09:55:02Z")

</div>

This is now implemented in the latest version of the wrapper, which is live already on Galaxy Europe, in case you want to test it.