MoTIF-X improves drug molecule prediction using motif-based learning
MoTIF-X: A Multimodal Tokenized Framework for Interpretable and Extensible Molecular Representation Learning
Machine Learning
Summary
Understanding how molecules behave helps design new medicines, but molecules are complex and look different depending on how you view them. The authors present MoTIF-X, a method that focuses on smaller parts of molecules called motifs to better combine different molecular views like graphs and 3D structures. This approach allows for more detailed and understandable predictions about molecule properties and how drugs interact with targets. Their method showed improved accuracy on multiple drug-related tests and can explain which parts of a molecule are most important for its activity.
What this means in practice
- •For drug discovery teams: Improve prediction of drug molecule properties by integrating multiple molecular views with motif-based learning to find promising candidates more accurately.
- •For pharmaceutical data teams: Enhance drug-target interaction models with a transferable framework that generalizes to unseen drugs without extra tuning, aiding early drug screening.
Authors
Linqing Mo, Jiayu Zhou, Bin Chen
Abstract
Molecular representation learning is central to computer-aided drug discovery. Molecular graphs, SMILES strings, and 3D conformations provide complementary structural information, yet many multimodal approaches encode these views independently and align them only at a later stage, limiting fine-grained cross-modal interaction and substructure-level interpretability. To address these limitations, we introduce MoTIF-X, a motif-centered framework that uses graph-grounded chemical motifs as shared anchors for multimodal integration and interpretation. Its first pretraining stage learns motif representations through hierarchical contrastive learning across atomic, motif, and molecular scales. The second stage contextualizes these representations with SMILES and torsion-angle tokens through multimodal masked token modeling. After pretraining on drug-like molecules with multiple conformers, MoTIF-X achieved the lowest mean absolute error on all nine OpenADMET ExpansionRx endpoints and the best overall performance among the evaluated methods. Significance analyses supported its advantage in the vast majority of endpoint-baseline comparisons after multiple-testing correction. Ablation studies supported the complementary contributions of motif-token contextualization, multimodal integration, and two-stage pretraining. Beyond molecular properties, the framework extended to drug-target interaction prediction, achieving the best average classification performance across the evaluated benchmarks and generalizing to an external drug-cold-start dataset without additional fine-tuning. Its motif-centered design also enabled substructure-level interpretation: higher motif attribution scores were associated with larger experimentally measured activity shifts. Together, these findings support MoTIF-X as a transferable and interpretable framework for molecular modeling.