Aligning Molecular Graph Explanations with Chemical Identity via InChIfied Invariants
This work addresses the problem of inconsistent predictions and explanations in molecular graph machine learning due to chemically equivalent but differently drawn graphs, which is crucial for reliable AI in chemistry and drug discovery.
The paper introduces InChIfied Invariants, a set of molecular graph features that are invariant under transformations preserving chemical identity, achieving 99.62% representation consistency for chemically equivalent graphs compared to 0.35% with standard Daylight invariants, while maintaining predictive performance and improving explanation consistency.
Obtaining consistent explanations for machine learning on molecular graphs requires predictions and attributions to be aligned with chemical identity. However, chemically equivalent drawings of the same molecule can induce different molecular representations, leading to inconsistent predictions and explanations. Here, we introduce InChIfied Invariants, a class of node, edge, and graph features based on the International Chemical Identifier (InChI) and designed to be invariant under transformations that preserve chemical identity. Using one million molecular graphs from PubChem Substances, we show that InChIfied Invariants produce identical representations for chemically equivalent graphs in 99.62% of cases, whereas standard Daylight invariants do so in only 0.35% of cases. Across MoleculeNet tasks, InChIfied Invariants preserve predictive performance while significantly improving prediction consistency across alternative graph depictions of the same molecules. We further perform a quantitative attribution analysis and show that explanations produced with standard molecular featurization methods vary substantially across chemically equivalent graphs, while InChIfied Invariants enforce consistent attributions by construction. We release open-source software implementing InChIfied Invariants, which can be used as a drop-in replacement for standard molecular graph features.