We believe science advances fastest when knowledge flows freely. VARL is committed to sharing the tools, formats, and research outputs that can safely accelerate biological discovery across the global scientific community.
What We're Building
A selection of our foundational tools and libraries are being prepared for open-source release. These include biological data preprocessing pipelines, molecular visualization components, standardized dataset formats, and reference implementations of our simulation protocols.
Development is active. Some repositories are already in private beta with select research partners. Public releases will be announced through our Latest page and distributed via GitHub under permissive licenses where possible, and restricted licenses where necessary.
varl-preprocess
Standardized pipelines for multi-omics data cleaning, normalization, and feature extraction. Designed to transform raw biological data into simulation-ready formats.
Python · Apache 2.0varl-viz
Interactive molecular and network visualization components for web applications. Built on WebGL with React bindings for embedding biological data visualizations into any interface.
TypeScript · MITvarl-schemas
Open data format specifications for digital twin models, simulation outputs, and biomarker datasets. Designed for interoperability across platforms and programming languages.
JSON Schema · CC BY 4.0Why Our Core Platform Is Not Open Source
We operate at the intersection of artificial intelligence and biological systems. This is not social media. This is not e-commerce. The tools we build can simulate disease mechanisms, design molecular interventions, and predict how living systems will respond to chemical compounds. In the wrong hands, these capabilities can be weaponized.
A digital twin engine capable of modeling pathogen behavior can also be used to engineer more dangerous pathogens. A drug discovery platform can be repurposed to design toxins. A biomarker detection system can be turned into a surveillance tool. These are not hypothetical risks. They are documented concerns raised by biosecurity researchers, defense agencies, and the scientific community at large.
We have an absolute responsibility to ensure that our most powerful technologies are not accessible without verification, oversight, and accountability. Open-sourcing the core VARL platform would be irresponsible. It would place capabilities of enormous consequence into an uncontrolled environment with no mechanism for preventing misuse.
Maintaining control over our most powerful tools is a prerequisite for deploying them responsibly. The open-source community will receive everything we can safely share. The rest stays behind access controls, audit logs, and institutional review.
Controlled Access via API
Researchers, institutions, and governments that require access to VARL's full capabilities can apply through our API Access Request process. Every application is reviewed individually by our governance team. We evaluate the applicant's identity, institutional affiliation, intended use case, data handling practices, and compliance posture before granting any level of access.
Approved applicants receive scoped API keys with permissions tailored to their specific needs. All usage is logged, auditable, and subject to rate limits. Access can be revoked at any time if our terms are violated or if we determine that continued access poses a risk.
This process is intentionally rigorous. It is designed to ensure that access is granted only to those whose work serves the advancement of biological understanding.
Request API Access →Our Commitment
Research
Every major scientific finding produced by VARL is published in peer-reviewed journals. Our methodologies, validation protocols, and benchmark results are made publicly available. We believe in open science even when we cannot offer open code. The knowledge we generate belongs to the scientific record.
Tooling
Every utility, library, and data format that can be safely open-sourced will be. We actively maintain open repositories and welcome community contributions. Our goal is to give the research community the best possible infrastructure for biological computation, within the boundaries of what is safe to release.
Responsibly
We will never use “safety” as an excuse to hoard technology for competitive advantage. Every restriction we place on access is driven by genuine risk assessment, not market positioning. We submit our access policies to independent review and publish annual transparency reports documenting every access decision we make.
Access
As governance frameworks mature and international biosecurity standards evolve, we intend to progressively open more of our platform. The goal is not permanent restriction. It is responsible, phased release — expanding access as the world builds the institutional capacity to handle these tools safely. We are working toward a future where more is open, not less.
The future of science depends on shared knowledge.
We intend to contribute to that future, carefully.