rosettacommons/foundry

How rosettacommons/foundry works

Foundry is a shared framework for biomolecular model training and inference, shipping four model families: RFD3 and RFD3NA for constrained all-atom design, RF3 for structure prediction, and ProteinMPNN/LigandMPNN for inverse folding. The user-facing workflow is to provide a structure or design specification, prepare model-specific inputs, run the model or sampler, and obtain designed sequences or predicted/generated structures. AtomWorks supplies structure manipulation and preprocessing across model workflows. The graph captures documented capabilities and indexed module references; several model internals and the RF3 engine were not excerpted, so those links are shown as unverified.

Main entry points: rfd3/engine.py Design Engine · rfd3na/engine.py Design Engine · rf3.py Inference Engine · mpnn.py Inference Engine

Components 27

Foundry Core

Inference Base src/foundry/inference_engines/base.py
Checkpoint Registry src/foundry/inference_engines/checkpoint_registry.py
Component Utilities src/foundry/utils/components.py
Fabric Training src/foundry/trainers/fabric.py

RFD3 Design

Input Preparation models/rfd3/src/rfd3/utils/inference.py
Conditioning Pipeline models/rfd3/src/rfd3/transforms/pipelines.py
Design Engine models/rfd3/src/rfd3/engine.py
Diffusion Sampler models/rfd3/src/rfd3/model/inference_sampler.py
Design Model
Designed Structures models/rfd3/src/rfd3/engine.py

RFD3NA Design

Input Preparation models/rfd3na/src/rfd3na/utils/inference.py
Conditioning Pipeline models/rfd3na/src/rfd3na/transforms/pipelines.py
Design Engine models/rfd3na/src/rfd3na/engine.py
Diffusion Sampler models/rfd3na/src/rfd3na/model/inference_sampler.py
Design Model models/rfd3na/src/rfd3na/model/RFD3.py
Designed Structures models/rfd3na/src/rfd3na/engine.py

RF3 Folding

Data Pipelines models/rf3/src/rf3/data/pipelines.py
Inference Engine models/rf3/src/rf3/inference_engines/rf3.py
Folding Model models/rf3/src/rf3/model/RF3.py
Predicted Structures models/rf3/src/rf3/utils/inference.py

MPNN Inverse Folding

Inference Inputs models/mpnn/src/mpnn/utils/inference.py
Inference Engine models/mpnn/src/mpnn/inference_engines/mpnn.py
Sequence Model models/mpnn/src/mpnn/model/mpnn.py
Graph Embeddings models/mpnn/src/mpnn/model/layers/graph_embeddings.py
Designed Sequences models/mpnn/src/mpnn/inference.py

Other components

Researcher
AtomWorks

Connections 29

Researcher → Design Engine: specifies design
Design Engine → Input Preparation: prepares inputs models/rfd3/src/rfd3/engine.py
Design Engine → Diffusion Sampler: runs sampler models/rfd3/src/rfd3/engine.py
Conditioning Pipeline → Input Preparation: uses conditioning models/rfd3/src/rfd3/transforms/pipelines.py
Diffusion Sampler → Design Model: denoises with
Design Engine → Designed Structures: writes outputs models/rfd3/src/rfd3/engine.py
Researcher → Design Engine: specifies design
Design Engine → Input Preparation: prepares inputs models/rfd3na/src/rfd3na/engine.py
Design Engine → Diffusion Sampler: runs sampler models/rfd3na/src/rfd3na/engine.py
Conditioning Pipeline → Input Preparation: uses conditioning models/rfd3na/src/rfd3na/transforms/pipelines.py
Diffusion Sampler → Design Model: denoises with
Design Engine → Designed Structures: writes outputs models/rfd3na/src/rfd3na/engine.py
Researcher → Inference Engine: requests folding
Inference Engine → Data Pipelines: loads data
Inference Engine → Folding Model: runs prediction
Inference Engine → Predicted Structures: returns structures
Researcher → Inference Inputs: provides backbone
Inference Inputs → Inference Engine: builds request models/mpnn/src/mpnn/inference.py
Inference Engine → Sequence Model: runs inference models/mpnn/src/mpnn/inference_engines/mpnn.py
Sequence Model → Graph Embeddings: uses embeddings models/mpnn/src/mpnn/model/mpnn.py
Inference Engine → Designed Sequences: returns sequences models/mpnn/src/mpnn/inference.py
AtomWorks → Input Preparation: processes structures
AtomWorks → Input Preparation: processes structures
AtomWorks → Data Pipelines: processes structures
AtomWorks → Inference Inputs: processes structures
Inference Base → Checkpoint Registry: looks up weights src/foundry/inference_engines/base.py
Fabric Training → Component Utilities: instantiates components src/foundry/trainers/fabric.py
Design Engine → Inference Base: extends engine models/rfd3/src/rfd3/engine.py
Design Engine → Inference Base: extends engine models/rfd3na/src/rfd3na/engine.py

More Python diagrams: scikit-hep/awkward · yuchenlin/llm-blender · azuread/microsoft-authentication-library-for-python · andrew6rant/andrew6rant · motional/nuplan-devkit

Drawn by GitDiagram on September 30, 2026 from the repository's file tree, README and source files. Read as Markdown ·