Lune

ICLR2025Top-tier venue

CL-MFAP: A Contrastive Learning-Based Multimodal Foundation Model for Molecular Property Prediction and Antibiotic Screening

Gen Zhou, Sugitha Janarthanan, Yutong Lu, Pingzhao Hu

Abstract

Due to the rise in antimicrobial resistance, identifying novel compounds with antibiotic potential is crucial for combatting this global health issue. However, traditional drug development methods are costly and inefficient. Recognizing the pressing need for more effective solutions, researchers have turned to machine learning techniques to streamline the prediction and development of novel antibiotic compounds. While foundation models have shown promise in antibiotic discovery, current mainstream efforts still fall short of fully leveraging the potential of multimodal molecular data. Recent studies suggest that contrastive learning frameworks utilizing multimodal data exhibit excellent performance in representation learning across various domains. Building upon this, we introduce CL-MFAP, an unsupervised contrastive learning (CL)-based multimodal foundation (MF) model specifically tailored for discovering small molecules with potential antibiotic properties (AP) using three types of molecular data. This model employs 1.6 million bioactive molecules with drug-like properties from the ChEMBL dataset to jointly pretrain three encoders: (1) a transformer-based encoder with rotary position embedding for processing SMILES strings; (2) another transformerbased encoder, incorporating a novel bi-level routing attention mechanism to handle molecular graph representations; and (3) a Morgan fingerprint encoder using a multilayer perceptron, to achieve the contrastive learning purpose. The CL-MFAP outperforms baseline models in antibiotic property prediction by effectively utilizing different molecular modalities and demonstrates superior domain-specific performance when fine-tuned for antibiotic-related property prediction tasks.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

Questions to start from

Lune can draw on

the full paper · related work

Get the answer

Free to start. No credit card required.

Lineage

Builds onCited byLeft to right by year; the more cited, the higher.

Builds on4

  1. Graph Contrastive Learning with AugmentationsYuning You, Tianlong Chen, Yongduo Sui, Ting Chen et al.NeurIPS 2020 · 3,042 citations
  2. On the Connection Between MPNN and Graph TransformerChen Cai, Truong Son Hy, Rose Yu, Yusu WangICML 2023 · 82 citations
  3. UniCorn: A Unified Contrastive Learning Approach for Multi-view Molecular Representation LearningShikun Feng, Yuyan Ni, Minghao Li, Yanwen Huang et al.ICML 2024 · 22 citations
  4. ParameterNet: Parameters are All You Need for Large-Scale Visual Pretraining of Mobile NetworksKai Han, Yunhe Wang, Jianyuan Guo, Enhua WuCVPR 2024

Related papers

  1. GRAM-DTI: Adaptive Multimodal Representation Learning for Drug–Target Interaction PredictionFeng Jiang, Amina Mollaysa, Hehuan Ma, Yuzhi Guo et al.ICLR 2026
  2. Molecular Contrastive Learning with Chemical Element Knowledge GraphYin Fang, Qiang Zhang, Haihong Yang, Xiang Zhuang et al.AAAI 2022 · 140 citations
  3. Fragment-based Pretraining and Finetuning on Molecular GraphsKha-Dinh Luong, Ambuj K. SinghNeurIPS 2023 · 36 citations
  4. Multi-modal contrastive learning adapts to intrinsic dimensions of shared latent variablesYu Gui, Cong Ma, Zongming MaNeurIPS 2025 · 9 citations
  5. Drugging the Undruggable: Benchmarking and Modeling Fragment-Based ScreeningHaichuan Tan, Bowen Gao, Jiaxin Li, Yinjun Jia et al.ICLR 2026
Dusk over the sea between two cliffs drawn in fine vertical lines