All model cards

Cytotoxicity

classification · XGBoost (classification)

Purpose

Predicts general cytotoxicity (cell damage / reduced cell viability) potential.

Prediction Output

Binary classification (cytotoxic / non-cytotoxic) + probability.

Input Requirements

A single valid SMILES string, standardized server-side before featurization.

Training Methodology

Algorithm: XGBoost (classification) · Optuna, TPE sampler, 100 trials, random_state=42

Feature representation: 2,048-bit Morgan ECFP4 (radius=2) + 167 MACCS keys + 25 RDKit physicochemical descriptors → 2,240 raw features → variance threshold (>0.01) → Pearson correlation filter (|r|<0.95) → top 500 features → StandardScaler (fit on the training fold only).

  • Sources: literature + ChEMBL cytotoxicity assays. N = 387 compounds (as documented in reproducibility.md — the repo's own train/test split figures for this endpoint, 118/34, do not sum to 387; reported here verbatim rather than reconciled).

Validation Methodology

  • Scaffold-split GroupKFold (k=5).

Scaffold-split AUC = 0.916 — the highest of all seven endpoints.

Known Limitations

  • Applicability domain coverage on the training set: 100%.

Applicability Domain

Leverage-based applicability domain: h* = 3×(p+1)/n (Williams plot). in_domain=true → confidence 1.0; in_domain=false → confidence 0.0, and the prediction should be treated as exploratory only. Supplementary checks: Tanimoto k-NN distance > 0.6 or descriptor values outside [min−3σ, max+3σ].

Interpretability Support

SHAP (SHapley Additive exPlanations) feature attribution is available via /predict/interpret and /predict/integrated, returning per-feature contribution values and a global/local breakdown by feature category (physicochemical / structural fingerprints / substructure keys).

Research-Use Disclaimer

Research use only. Not validated for regulatory submissions or clinical decision-making. Predictions must not be the sole basis for safety decisions without independent experimental validation.