Adding scVIVA-Tools and scIB-Metrics to the ecosystem package registry

[ori-kron-wis]

scVIVA-Tools

Name of the tool: scviva-tools

Short description: A consolidated spatial transcriptomics toolkit built on scvi-tools, bundling niche-aware representation learning (scVIVA), deconvolution (DestVI, Stereoscope), noise correction (ResolVI), gene imputation (GIMVI), and spatial mapping (Tangram) under one package.

How does the package use scverse data structures (please describe in a few sentences): All models take AnnData as their primary input via the scvi-tools setup_anndata pattern, storing counts, batch/covariate annotations, and spatial coordinates in adata.obs/.obsm. The [spatial] extra adds interoperability with SpatialData and squidpy for coordinate-aware spatial workflows.

• The code is publicly available under an OSI-approved license (BSD-3-Clause)
• The package provides versioned releases (tagged releases on PyPI, currently v0.1.6)
• The package can be installed from a standard registry (PyPI: scviva-tools)
• Automated tests cover essential functions of the package and a reasonable range of inputs and conditions
• Continuous integration (CI) automatically executes these tests on each push or pull request (GitHub Actions: test.yaml, test_macos.yaml, test_cuda.yaml, test_optional.yaml)
• The package provides API documentation via a website or README (https://scviva-tools.readthedocs.io/)
• The package uses scverse datastructures where appropriate (AnnData, SpatialData)
• I am an author or maintainer of the tool and agree on listing the package on the scverse website
• I agree to abide by the scverse code of conduct on all scverse communication channels

Recommended

• Please announce this package on scverse communication channels (zulip, discourse, twitter)

• Please tag the author(s) these announcements. Handles (e.g. @scverse_team) to include are:

  • Zulip:
  • Discourse:
  • Mastodon:
  • Bluesky:
  • Twitter:

• The package provides tutorials (or "vignettes") that help getting users started quickly

• The package uses the scverse cookiecutter template.

• I would like to be invited to the scverse Github organization

---

scIB-Metrics

Name of the tool: scib-metrics

Short description: Accelerated, Python-only implementations of the scIB metrics for benchmarking single-cell data integration methods, using JAX for jit-compilation and hardware acceleration.

How does the package use scverse data structures (please describe in a few sentences): Metrics operate directly on AnnData objects, taking integration outputs (e.g. embeddings/PCA in adata.obsm) alongside batch and cell-type annotations in adata.obs as input to compute benchmarking scores.

• The code is publicly available under an OSI-approved license (BSD-3-Clause)
• The package provides versioned releases (tagged releases on PyPI, currently v0.6.0)
• The package can be installed from a standard registry (PyPI: scib-metrics)
• Automated tests cover essential functions of the package and a reasonable range of inputs and conditions
• Continuous integration (CI) automatically executes these tests on each push or pull request (GitHub Actions: test.yaml, test_macos.yaml, test_windows.yaml, test_linux_cuda.yaml)
• The package provides API documentation via a website or README (https://scib-metrics.readthedocs.io/)
• The package uses scverse datastructures where appropriate (AnnData)
• I am an author or maintainer of the tool and agree on listing the package on the scverse website
• I agree to abide by the scverse code of conduct on all scverse communication channels

Recommended

• Please announce this package on scverse communication channels (zulip, discourse, twitter)

• Please tag the author(s) these announcements. Handles (e.g. @scverse_team) to include are:

  • Zulip:
  • Discourse:
  • Mastodon:
  • Bluesky:
  • Twitter:

• The package provides tutorials (or "vignettes") that help getting users started quickly

• The package uses the scverse cookiecutter template.

• I would like to be invited to the scverse Github organization

[Zethson]

Will have look soon.

[grst]

Should scviva-tools become a core package (maybe not immediately)? After all it contains a couple of models that were previously in scvi-tools.

[ori-kron-wis]

I think yes, but not right now - maybe towards the end of year or so.

Only after all our planned models have been added to it and it will gain some user traffic.
Once that happens, we will also remove the overlapping models from scvi-tools (which also has a few new things in the pipe by itself)