Leveraging evolution to engineer biology

We're testing an evolution-based drug discovery model
with an iterative loop of computational prediction
and experimental validation.

Target identification

Mining billions of years of evolution to systematize target discovery

We integrate across evolutionary scales to identify the causal bases of biological traits.

Model selection

Improving outcomes by picking useful research organisms

We're applying our platform to match disease biology with the most human relevant organisms.

Biologics design

Designing proteins from evolutionary principles

We develop statistical models to predict fitness and function. Our closed loop lets us experimentally test predictions and iterate on our models.

Technology development

Building assays and tools to test our predictions

We're developing and deploying organism-agnostic tools to investigate disease mechanisms.

Target identification

Mining billions of years of evolution to systematize target discovery

We integrate across evolutionary scales to identify the causal bases of biological traits.

Model selection

Improving outcomes by picking useful research organisms

We're applying our platform to match disease biology with the most human relevant organisms.

Biologics design

Designing proteins from evolutionary principles

We develop statistical models to predict fitness and function. Our closed loop lets us experimentally test predictions and iterate on our models.

Technology development

Building assays and tools to test our predictions

We're developing and deploying organism-agnostic tools to investigate disease mechanisms.

We publish our work and release open-source tools and resources for the scientific community.

Use our tools

Check out our latest research

resource10 Jun 2026

An automated workflow for kit-based ELISAs

We developed a workflow to run the Cayman His-tag detection ELISA kit on the Opentrons OT-2 liquid-handling robot. We encourage others to adapt these resources for other kits.

Read now

result04 Jun 2026

Interpretable neural networks reveal global epistasis in yeast

We applied a mechanistic interpretability approach to neural networks trained on a large yeast genotype–phenotype dataset. This allowed us to uncover novel evidence of environmentally mediated global epistasis in these data.

Read now

resource03 Jun 2026

Agentic PyMOL: Structural analysis in PyMOL through natural language

Agentic PyMOL is an MCP server that gives language-model agents discoverable, two-way access to a user's live PyMOL session, letting agents both manipulate molecular visualizations and read structured analytical results back out.

Read now