# Designing proteins with purpose.

RFdiffusion enabled researchers to generate proteins with specified structural and functional goals.

Canonical: https://brightaifuture.com/discoveries/rfdiffusion
Format: story
Source publication: 2023-07-11
Bright publication: 2026-09-05
Substantive update: None recorded
Evidence and review: Demonstrated; confidence: unassessed; source-checked; legacy source check. Legacy source check; no named human reviewer is recorded in this projection.

## The human problem

Useful protein structures are difficult to design from scratch.

## The prior constraint

Design requires searching a huge space of structures and sequences.

## AI’s actual role

A diffusion model generated candidate protein backbones.

## The documented result

The study included computational benchmarks and experimental tests of designs.

## Why it may matter

New tools could support biological research and future therapeutic development.

## Limitations

A designed binder is not an approved medicine; each application needs validation.

## Unresolved questions

Investigate function, safety, and reliability in intended applications.

## Provenance and history

{
  "dates": {
    "eventDate": null,
    "publicationDate": "2023-07-11",
    "captureDate": null,
    "lastReviewedDate": "2026-09-05"
  },
  "provenance": {
    "origin": "legacy-projection",
    "externalId": "rfdiffusion"
  },
  "revisions": [],
  "corrections": []
}

## Original sources

- [De novo design of protein structure and function with RFdiffusion](https://www.nature.com/articles/s41586-023-06415-8)

## Continue exploring

- [Concise evidence record](https://brightaifuture.com/discoveries/rfdiffusion)