Companies GUIDE

Recursion Pharmaceuticals AI

Recursion Pharmaceuticals runs one of the world's largest automated biology labs, generating petabytes of cellular images to let machine learning models map how drugs change cells.

2 min readLast updated

Overview

It matters because it turns wet-lab biology into a data problem that AI can search at industrial scale.

Deep Dive

Recursion, founded in 2013 and based in Salt Lake City, built its strategy around 'phenomics' — taking microscope images of human cells treated with thousands of compounds and genetic perturbations, then using deep learning to convert each image into a numerical fingerprint. Cells with similar fingerprints likely share biology, so a disease-altered cell that a drug pushes back toward 'healthy' becomes a candidate hit. Its robotic labs run millions of experiments weekly, feeding the Recursion Operating System (now branded under the merged Recursion-Exscientia company). In 2023 NVIDIA invested $50 million, and Recursion released the open BioHive supercomputer and large datasets like RxRx3. The approach trades hand-picked targets for unbiased, data-driven discovery across many diseases at once.

Technical Insight

Recursion uses Cell Painting: cells are stained with fluorescent dyes marking organelles like the nucleus, mitochondria, and cytoskeleton, then imaged across channels. Convolutional and increasingly transformer-based models embed each image into a high-dimensional vector. Crucially, the team applies heavy batch-correction to remove technical artifacts (plate, day, instrument) so that biological signal dominates. Drugs are ranked by how their embeddings shift diseased cells toward healthy reference states.

Strategic Impact

Vendor strategy

Vendor roadmaps influence what features your team can build next.

Cost and budget

Commercial terms and deployment options affect long-term cost and risk.

Risk and safety

Company incentives shape product defaults, safety posture, and openness.

The Future of Recursion Pharmaceuticals AI

After merging with Exscientia in 2024, Recursion is combining its phenomics maps with structure-based chemistry design, aiming to compress discovery timelines and cost. Expect bigger foundation models trained on its proprietary maps of biology and chemistry, more partnered programs with Roche, Genentech, Sanofi and Bayer, and growing scrutiny on whether AI-originated candidates succeed in human clinical trials, the real test of the platform.

Real-World Implementation

Screening tens of thousands of compounds against cells modeling rare genetic diseases like cerebral cavernous malformation, advancing candidates such as REC-994 into trials.

Using Cell Painting phenotypes to repurpose existing drugs for new indications by spotting unexpected cellular similarities.

Releasing the RxRx3 public dataset of millions of cell images so outside researchers can train and benchmark biology models.

Partnering with Roche and Genentech to map neuroscience and gastrointestinal cancer biology at industrial scale.

Risks & Guardrails

Launch announcements may outpace stability in real production workflows.

API pricing or policy shifts can break assumptions overnight.

Single-vendor dependency increases lock-in and migration costs.

Implementation Roadmap

1

Evaluate providers using your own tasks and datasets.

2

Review privacy, security, and legal terms before integration.

3

Maintain a fallback plan across models or vendors.

4

Monitor release notes so roadmap changes do not surprise teams.

Keep Exploring

Free newsletter

Get the daily AI briefing

Three verified AI stories every weekday morning, written in plain English. Free forever, no ads.

One email each weekday. Unsubscribe in one click. We never sell or share your address.

Test yourself

Take the Recursion Pharmaceuticals AI quiz

Instant feedback on every answer, and a shareable certificate with a verifiable ID once you pass a course.

Start quiz

Support free AI education. AI Understanding is a 501(c)(3) nonprofit — no ads, no paywall, ever. Make a donation

Next guide

Insitro Machine Learning Biology

Frequently asked questions

What is Recursion Pharmaceuticals AI?

Recursion Pharmaceuticals runs one of the world's largest automated biology labs, generating petabytes of cellular images to let machine learning models map how drugs change cells. It matters because it turns wet-lab biology into a data problem that AI can search at industrial scale.

What is the core data type Recursion uses to train its AI models?

Recursion's 'phenomics' approach centers on images of cells treated with drugs or genetic changes, turned into numerical fingerprints.

What does the Cell Painting technique do?

Cell Painting uses fluorescent dyes to highlight structures like the nucleus and mitochondria, captured across imaging channels.

Why does Recursion apply heavy batch-correction to its data?

Differences in plate, day, or instrument can swamp real biology, so batch-correction isolates the genuine biological signal.

Which company invested $50 million in Recursion in 2023?

NVIDIA invested $50 million in 2023, reflecting the GPU-heavy nature of Recursion's image-based AI.

In 2024, Recursion merged with which AI drug-discovery company?

Recursion merged with Exscientia, combining phenomics with structure-based chemistry design.