How zhenjiagan/cateso works
CatESO predicts enzyme kcat from protein embeddings and substrate molecular graphs, and optimizes enzyme sequences against that prediction objective. Its main paths are embedding generation, ensemble training/inference, and differentiable sequence optimization with optional ESMFold structural constraints. The graph also shows the model and structure-prediction layers. Relationships are based on README statements and SOURCE INDEX references; unsampled implementation details are not expanded.
Main entry points: models.py CatESO Model · esmfold.py ESMFold · main_ensemble.py Ensemble Training · inference.py Ensemble Inference
Components 21
kcat Prediction
- Ensemble Training
main_ensemble.py - Ensemble Inference
inference.py - Prediction Dataset
run/dataloader.py - Training Runtime
run/trainer.py
Protein Embeddings
- Embedding Generation
scripts/embedding.py - ESM-2
esm/model/esm2.py - Sequence Batching
esm/data.py
Sequence Optimization
- Optimization Entry
main_optimize.py - Sequence Optimizer
optimizer/sequence_optimizer.py - ESM-2 Manager
optimizer/esm_manager.py - Fold Evaluator
optimizer/esmfold_evaluator.py
Structure Modeling
- ESMFold
esm/esmfold/v1/esmfold.py - Folding Trunk
esm/esmfold/v1/trunk.py - Structure Module
openfold/openfold/model/structure_module.py - Protein Structures
openfold/openfold/np/protein.py - Residue Constants
openfold/openfold/np/residue_constants.py
Model Internals
- CatESO Model
models.py - Substrate Fusion
module/CN.py - Protein Encoder
module/Encoder.py - Transformer Layers
module/Transformer.py
Other components
- Researcher
Connections 23
- Researcher → Ensemble Training: trains
- Researcher → Ensemble Inference: requests prediction
- Researcher → Optimization Entry: starts optimization
- Embedding Generation → ESM-2: embeds sequences
- Embedding Generation → Sequence Batching: batches sequences
- Ensemble Training → CatESO Model: trains
- Ensemble Training → Training Runtime: uses runtime
- Ensemble Inference → Prediction Dataset: loads rows
inference.py - Ensemble Inference → CatESO Model: predicts with
inference.py - Prediction Dataset → CatESO Model: supplies inputs
- CatESO Model → Substrate Fusion: uses
models.py - CatESO Model → Protein Encoder: uses
models.py - Substrate Fusion → Transformer Layers: uses
module/CN.py - Optimization Entry → Sequence Optimizer: runs
main_optimize.py - Optimization Entry → ESM-2 Manager: loads ESM-2
main_optimize.py - Sequence Optimizer → CatESO Model: optimizes against
- Sequence Optimizer → Fold Evaluator: requests constraint
optimizer/sequence_optimizer.py - Fold Evaluator → ESMFold: evaluates folds
- ESMFold → Folding Trunk: uses trunk
esm/esmfold/v1/esmfold.py - Folding Trunk → Structure Module: uses structure module
esm/esmfold/v1/trunk.py - ESMFold → Residue Constants: uses residue definitions
esm/esmfold/v1/esmfold.py - ESMFold → Protein Structures: uses protein structures
- ESM-2 Manager → ESM-2: loads model
optimizer/esm_manager.py