All model cards
Ames Mutagenicity
classification · XGBoost (classification)
Purpose
Predicts Ames test mutagenicity (bacterial reverse mutation assay) potential — a standard genotoxicity screen (ICH S2 relevant).
Prediction Output
Binary classification (mutagenic / non-mutagenic) + 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: curated literature + ChEMBL v34. N = 876 compounds in the full curated set.
- The deployed model trains on an active subset of the 876; exact inclusion/exclusion criteria are documented in data/metadata/ames_manifest.json in the backend repo. Expansion to the full 876-compound set is a documented roadmap item.
Validation Methodology
- Scaffold-split GroupKFold (k=5).
Scaffold-split AUC = 0.852.
Known Limitations
- Training uses a subset of the full curated 876-compound dataset, not all of it — see data/metadata/ames_manifest.json for the exact criteria.
- 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.