Research map / Label-efficient and robust learning / Recognition and long-tail
GRAM-DTI: Adaptive Multimodal Representation Learning for Drug–Target Interaction Prediction
Feng Jiang, Amina Mollaysa, Hehuan Ma, Yuzhi Guo, Tommaso Mansi, Junzhou Huang, Mangal Prakash, Rui Liao
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Gramian volume-based four-modality alignment (ablation) | GRAM-DTI · Activation and Yamanishi_08 ablation | AUROC/AUPRC etc. | Improved |
| add IC50 auxiliary supervision (ablation) | GRAM-DTI · Activation ablation | AUROC/AUPRC etc. | Improved |
| add IC50 supervision in Activation target cold start (ablation) | GRAM-DTI · Activation target cold start | AUROC/AUPRC etc. | Hurt |
| add bimodal SMILES-protein contrastive loss (ablation) | GRAM-DTI · Activation ablation | AUROC/AUPRC etc. | Improved |
| add gradient-informed adaptive modality dropout (ablation) | GRAM-DTI · Activation and Yamanishi_08 ablation | AUROC/AUPRC etc. | Improved |
| pre-train with GRAM-DTI instead of DTIAM | GRAM-DTI · Activation target cold start | AUPR: 0.391 -> 0.450 | Improved |
| pre-train with GRAM-DTI instead of DTIAM | GRAM-DTI · Yamanishi_08 warm start | AUPR: 0.901 -> 0.904 | Improved |
| use GRAM-DTI instead of AI-DTI | GRAM-DTI · Inhibition warm start | AUPR: 0.845 -> 0.785 | Hurt |
| use GRAM-DTI instead of DTIAM | GRAM-DTI · Hetionet warm start | AUPR: 0.879 -> 0.859 | Hurt |
| use GRAM-DTI representations for zero-shot retrieval instead of DTIAM | GRAM-DTI · Yamanishi_08 S->P | Recall@10: 0.0341 -> 0.1795 | Improved |
| use GRAM-DTI for P->S retrieval on Hetionet | GRAM-DTI · Hetionet P->S | Recall@1: 0.0404 -> 0.0238 | Hurt |
Closest papers by meaning
- R-DTI: Drug Target Interaction Prediction Based on Second-Order Relevance Exploration (AAAI 2025 · similarity 0.89)
- Knowledge Enhanced Representation Learning for Drug Discovery (AAAI 2024 · similarity 0.88)
- Blend the Separated: Mixture of Synergistic Experts for Data-Scarcity Drug-Target Interaction Prediction (AAAI 2025 · similarity 0.87)
- Multilevel Attention Network with Semi-supervised Domain Adaptation for Drug-Target Prediction (AAAI 2024 · similarity 0.87)
- Omni-Mol: Multitask Molecular Model for Any-to-any Modalities (NeurIPS 2025 · similarity 0.86)
- Multimodal Molecular Pretraining via Modality Blending (ICLR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
