GenomicsXAI is a shared, unaffiliated GitHub organization dedicated to advancing the field of genomics-informed machine learning and explainable AI (XAI).

Our goal is to provide a neutral, community-driven space where researchers can collaboratively develop, maintain, and improve open tooling, benchmarks, and infrastructure relevant to genomics models and interpretability research to understand the cis-regulatory code.

This organization is not affiliated with any single lab, institution, or company.

The organization exists to:

• Consolidate reusable infrastructure for genomics ML and XAI.
• Enable community-driven development of fine-tuning, evaluation, and interpretability tools.
• Reduce duplicated engineering effort across labs.
• Promote reproducibility and transparent benchmarking.
• Encourage cross-framework interoperability (e.g., JAX, PyTorch).
• We prioritize practical tooling that benefits the broader research community.

All repositories under this organization are intended to be:

• Community-maintained
• Open to pull requests
• Designed for extensibility and reproducibility

We encourage:

• Bug fixes
• Performance improvements
• New training strategies
• Improved evaluation pipelines
• Documentation enhancements
• Cross-framework ports
• Reproducibility improvements

Contributors should:

• Use clear, modular design patterns.
• Include documentation and minimal working examples.
• Provide reproducible training or evaluation scripts.
• Add unit tests where possible.

Each repository may define additional contribution guidelines specific to its scope.

If you would like to contribute a repository to this organization:

• Transfer your existing repo into this org or
• Open a pull request to add a new repository

In your PR, briefly describe:

• What the repository does
• Who will help maintain it

The license

• Repositories should provide clear value to the genomics ML / XAI community and be open to community contributions.
• If you are unsure whether your project is a good fit, open a draft PR and start the discussion there.

This organization follows a lightweight governance model:

• Repository maintainers are responsible for reviewing and merging PRs.
• Maintainers are expected to act in the interest of the broader community.
• No single lab controls the direction of the organization.
• Major architectural changes should be discussed via issues before implementation.
• If a repository becomes inactive, additional maintainers may be added.

Some repositories may initially be private (e.g., due to ongoing development or licensing constraints).

If access is required:

• Add your GitHub username in the repository settings (if enabled), or
• Open an issue requesting access.

The long-term goal is to make infrastructure broadly accessible whenever possible.

Each repository defines its own license. We strongly encourage permissive open-source licenses to maximize reuse and collaboration.

If you use tools developed under this organization:

• Cite the associated papers where applicable.
• Acknowledge the repository and contributors in derivative work.

Community credit and transparency are core values.

• Contact

For general discussion, suggestions, or coordination:

• Open an issue in this repository.
• Use repository-specific issues for technical matters.

Genomics foundation models and XAI tooling will benefit from shared engineering infrastructure. This organization exists to make that collaboration practical and sustainable.

We welcome contributions!