From SMILES to Enhanced Molecular Property Prediction: A Unified Multimodal Framework with Predicted 3D Conformers and Contrastive Learning Techniques
- 6 December 2024
- journal article
- research article
- Published by American Chemical Society (ACS) in Journal of Chemical Information and Modeling
- Vol. 64 (24) , 9173-9195
Abstract
We present a novel molecular property prediction framework that requires only the SMILES format as input but is designed to be multimodal by incorporating predicted 3D conformer representations. Our model captures comprehensive molecular features by leveraging both the sequential character structure of SMILES and the three-dimensional spatial structure of conformers. The framework employs contrastive learning techniques, utilizing InfoNCE loss to align SMILES and conformer embeddings, along with task-specific loss functions, such as ConR for regression and SupCon for classification. To address data imbalance, we incorporate feature distribution smoothing (FDS), a common challenge in drug discovery. We evaluated the framework through multiple case studies, including SARS-CoV-2 drug docking score prediction, molecular property prediction using MoleculeNet data sets, and kinase inhibitor prediction for JAK-1, JAK-2, and MAPK-14 using custom data sets curated from PubChem. The results consistently outperformed state-of-the-art methods, with ConR and FDS significantly improving regression tasks and SupCon enhancing classification performance. These findings highlight the flexibility and robustness of our multimodal model, demonstrating its effectiveness across diverse molecular property prediction tasks, with promising applications in drug discovery and molecular analysis.Keywords
Funding Information
- Ministry of Business, Innovation and Employment (RTVU2301)
This publication has 59 references indexed in Scilit:
- ResGAT: Residual Graph Attention Networks for molecular property predictionMemetic Computing, 2024
- Multitask Learning on Graph Convolutional Residual Neural Networks for Screening of Multitarget Anticancer CompoundsJournal of Chemical Information and Modeling, 2024
- Molecular representations in bio-cheminformaticsMemetic Computing, 2024
- Predicting Cardiotoxicity of Molecules Using Attention-Based Graph Neural NetworksJournal of Chemical Information and Modeling, 2024
- Limits and Prospects of Molecular Fingerprinting for Phenotyping Biological Systems Revealed through In Silico ModelingAnalytical Chemistry, 2023
- HiGNN: A Hierarchical Informative Graph Neural Network for Molecular Property Prediction Equipped with Feature-Wise AttentionJournal of Chemical Information and Modeling, 2022
- iCYP-MFE: Identifying Human Cytochrome P450 Inhibitors Using Multitask Learning and Molecular Fingerprint-Embedded EncodingJournal of Chemical Information and Modeling, 2021
- Analysis of therapeutic targets for SARS-CoV-2 and discovery of potential drugs by computational methodsActa Pharmaceutica Sinica B, 2020
- Analyzing Learned Molecular Representations for Property PredictionJournal of Chemical Information and Modeling, 2019
- Forging the Basis for Developing Protein–Ligand Interaction Scoring FunctionsAccounts of Chemical Research, 2017