yigang-chen/scope-dti

How yigang-chen/scope-dti works

SCOPE-DTI trains and evaluates a drug–target interaction predictor. The principal workflow loads configured train/validation/test interaction tables, prepares drug and protein features in either 3D graph or 1D sequence mode, batches them into the model, and trains/evaluates interaction scores. The graph preserves both protein-processing branches, drug representation, bilinear fusion, and evaluation outputs. Edges follow shown imports/calls or README descriptions; some internal model dispatch and dataset-construction details are unsampled, so those relationships are omitted rather than inferred.

Main entry points: main.py Experiment runner · models.py SCOPE predictor · dataloader.py DTI datasets · trainer.py Trainer and evaluation

Components 17

Run orchestration

Configuration configs.py
Experiment runner main.py

Data preparation

Interaction splits data/demo_data
Protein coordinates data/demo_unip_cords.pkl
DTI datasets dataloader.py
Protein graph builder graph_utils.py
Feature encoding and batching utils.py

Prediction model

Training and results

Trainer and evaluation trainer.py
Metrics and checkpoints trainer.py

Other components

Researcher

Connections 17

Researcher → Experiment runner: runs README.md
Experiment runner → Configuration: loads main.py
Experiment runner → Interaction splits: reads splits main.py
Experiment runner → DTI datasets: builds datasets main.py
DTI datasets → Protein graph builder: builds graphs dataloader.py
DTI datasets → Protein coordinates: reads coordinates dataloader.py
Experiment runner → Feature encoding and batching: selects collator main.py
Experiment runner → SCOPE predictor: constructs main.py
SCOPE predictor → Drug GVP encoder: encodes drug models.py
SCOPE predictor → Protein graph encoder: encodes graph models.py
SCOPE predictor → Protein sequence CNN: encodes sequence models.py
Drug GVP encoder → GVP layers: uses layers models.py
SCOPE predictor → Bilinear attention fusion: fuses features models.py
Bilinear attention fusion → Interaction classifier: scores interaction models.py
Experiment runner → Trainer and evaluation: starts training main.py
Trainer and evaluation → SCOPE predictor: trains and evaluates trainer.py
Trainer and evaluation → Metrics and checkpoints: writes metrics trainer.py

More Python diagrams: xuejifang/lewam · yankeexe/llm-function-calling-demo · yanlinf/casent · ynput/ayon-maya · yourlabs/djwebdapp

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

How yigang-chen/scope-dti works

Main entry points: main.py Experiment runner · models.py SCOPE predictor · dataloader.py DTI datasets · trainer.py Trainer and evaluation

Connections 17

Researcher → Experiment runner: runs README.md
Experiment runner → Configuration: loads main.py
Experiment runner → Interaction splits: reads splits main.py
Experiment runner → DTI datasets: builds datasets main.py
DTI datasets → Protein graph builder: builds graphs dataloader.py
DTI datasets → Protein coordinates: reads coordinates dataloader.py
Experiment runner → Feature encoding and batching: selects collator main.py
Experiment runner → SCOPE predictor: constructs main.py
SCOPE predictor → Drug GVP encoder: encodes drug models.py
SCOPE predictor → Protein graph encoder: encodes graph models.py
SCOPE predictor → Protein sequence CNN: encodes sequence models.py
Drug GVP encoder → GVP layers: uses layers models.py
SCOPE predictor → Bilinear attention fusion: fuses features models.py
Bilinear attention fusion → Interaction classifier: scores interaction models.py
Experiment runner → Trainer and evaluation: starts training main.py
Trainer and evaluation → SCOPE predictor: trains and evaluates trainer.py
Trainer and evaluation → Metrics and checkpoints: writes metrics trainer.py

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