ailab-bio/PROTAC-Splitter-XGBoost

PROTAC-Splitter XGBoost

XGBoost graph edge classifier used by PROTAC-Splitter to split PROTAC molecules (heterobifunctional degraders) into their E3 ligand, linker, and POI warhead substructures.

Given a PROTAC's molecular graph, the model scores each candidate bond ("edge") for how likely it is to be one of the two cleavage points that separate the three substructures.

Usage

from protac_splitter import split_protac

result = split_protac("<PROTAC SMILES>", model="xgboost")

split_protac() auto-downloads and caches the model. It can also be loaded directly:

from protac_splitter.graphs.edge_classifier import GraphEdgeClassifier

model = GraphEdgeClassifier.from_pretrained("ailab-bio/PROTAC-Splitter-XGBoost")

Files

• PROTAC-Splitter-XGBoost.joblib — joblib-serialized GraphEdgeClassifier (scikit-learn pipeline wrapping XGBoost).
• manifest.json — model metadata: hyperparameters, feature lists, and a SHA-256 checksum of the model file.

Data

Trained on the curated PROTAC-Splitter dataset: https://doi.org/10.5281/zenodo.15797309