RosettaCommons/RoseTTAFold

Health v2 · maintenance only

Adoption not part of the score

2258 stars · 459 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

RoseTTAFold is the official implementation of a deep learning system for predicting protein structures and interactions using a three-track neural network. It provides Python models, scripts, and pretrained weights for monomer structure prediction, complex modeling, and protein-protein interaction screening.

Use cases

• predict the 3D structure of a protein from its amino acid sequence
• model protein-protein complexes and interactions
• screen for protein-protein interactions in yeast using a fast 2-track model
• run structure prediction pipelines comparable to AlphaFold
• fold proteins using MSA and template information
• estimate model confidence and accuracy with DeepAccNet

When to choose

• you need to predict protein structures from sequence with a published, well-cited method
• you want to model protein complexes or screen PPIs computationally
• you have Linux with NVIDIA GPUs and can download large sequence/structure databases
• you want an open-source (MIT code) alternative to AlphaFold with Rosetta integration

When to avoid

• you need a lightweight tool without large database downloads (BFD is 272G, templates 100G+)
• you need commercial use of the trained weights (they are non-commercial under the Rosetta-DL license)
• you work on Windows or macOS without GPU support
• you need a maintained turnkey web service rather than a research codebase

Facets

library · maturity stable

deep-learning machine-learning sdk bioinformatics deep-learning machine-learning python protein-structure-prediction protein-folding alphafold-alternative computational-biology pytorch complex-modeling ppi-screening linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="RosettaCommons/RoseTTAFold")