How hparklab/ifum works
IFUM predicts protein folding stability (ΔG) from sequence or folded structures. Its main workflow accepts FASTA or PDB/CIF input, predicts structures for FASTA with ESMFold, creates ProtT5 and ESM-IF1 embeddings, packages those with folded-state coordinates, runs the pretrained IEFFEUM model, and writes CSV predictions; optional per-residue output annotates structures. A separate training workflow prepares embeddings and trains the model. The sampled sources show the inference helpers, dataset, model, training, and visualization code; some internal calls in the unsampled portions of predict.py are therefore supported by its documented workflow rather than inspected implementation.
Main entry points: predict.py Prediction driver · model.py IEFFEUM model · utils.py Inference helpers · train.py Model trainer
Components 20
Prediction workflow
- Prediction driver
scripts/predict.py - ΔG CSV
Structure and features
- ESMFold
- Structure embeddings
scripts/predict.py - Sequence embeddings
scripts/predict.py - Prepared inputs
scripts/predict.py
Inference runtime
- Embedding dataset
src/IEFFEUM/dataset.py - Batch collation
src/IEFFEUM/dataset.py - Inference helpers
src/IEFFEUM/utils.py - IEFFEUM model
src/IEFFEUM/model.py - Model parameters
weights/params.pth
Training
- Training embeddings
scripts/training/prepare_embeddings.py - Training data
- Model trainer
scripts/training/train.py - Unfolded ensemble
src/IEFFEUM/flory.py
Per-residue interpretation
- Residue annotation
scripts/predict.py - Structure coloring
scripts/per_resi_dG.py
Other components
- Protein researcher
- FASTA sequences
- PDB/CIF structures
Connections 22
- Protein researcher → Prediction driver: runs
README.md - FASTA sequences → Prediction driver: provides sequence
README.md - PDB/CIF structures → Prediction driver: provides structures
README.md - Prediction driver → ESMFold: predicts structures
README.md - Prediction driver → Structure embeddings: extracts embeddings
scripts/predict.py - Prediction driver → Sequence embeddings: generates embeddings
scripts/predict.py - Prediction driver → Prepared inputs: packages features
scripts/predict.py - Prepared inputs → Embedding dataset: loads files
- Embedding dataset → Batch collation: pads batches
src/IEFFEUM/dataset.py - Prediction driver → Inference helpers: loads inference
scripts/predict.py - Inference helpers → Embedding dataset: constructs dataset
src/IEFFEUM/utils.py - Inference helpers → IEFFEUM model: loads model
src/IEFFEUM/utils.py - Model parameters → IEFFEUM model: parameter state
- Prediction driver → IEFFEUM model: runs prediction
README.md - Prediction driver → ΔG CSV: writes results
README.md - Protein researcher → Training embeddings: prepares training
scripts/training/prepare_embeddings.py - Training embeddings → Training data: saves embeddings
scripts/training/prepare_embeddings.py - Training data → Model trainer: loads examples
scripts/training/train.py - Model trainer → IEFFEUM model: trains model
scripts/training/train.py - Model trainer → Unfolded ensemble: builds ensemble
- Prediction driver → Residue annotation: optionally annotates
scripts/predict.py - Residue annotation → Structure coloring: writes annotations