MolLedger: An Additive Graph Neural Network with Chemically Grounded ADME Attributions
2026-08-31 • Machine Learning
Machine Learning
AI summaryⓘ
The authors developed a new type of machine learning model called MolLedger to predict certain drug-related properties of molecules. Instead of just giving a single prediction, MolLedger breaks it down into scores for each atom in the molecule, making it easier to understand why the model made its prediction. Their approach keeps the prediction accuracy high while providing clearer explanations linked to chemical properties. When compared to other methods, MolLedger gave better and more meaningful insights into how small changes in molecules affect their properties.
ADMEgraph neural networkmolecular property predictionmodel interpretabilityper-atom attributionmachine learningsmall molecule drug discoveryauxiliary lossmolecular graphexplainable AI
Authors
Christina X. Ji
Abstract
Optimizing absorption, distribution, metabolism, and excretion (ADME) is an important part of small molecule drug discovery. Many machine learning models have been built to predict ADME properties to facilitate this optimization process, but explaining model predictions is challenging. We propose a new graph neural network architecture with built-in meaningful per-atom attributions. Our model MolLedger outputs predictions that are the sum of per-atom scores. MolLedger's additive framework obtains exact interpretability at no cost to performance because the global context vector gives the additive head enough context to produce good per-atom scores. Furthermore, MolLedger produces attributions that are more faithful to chemical properties than other interpretability methods because the auxiliary loss in MolLedger anchors the atom scores to chemical properties. Our case studies comparing interpretations from multiple methods on molecular pairs reveal that MolLedger is much better at producing sensible explanations for predicted property changes.