---
title: "MultiGeo: Predicting Drug-Target Affinity via Adaptive Multi-Conformation Ensemble Learning"
authors:
  - "Ruida Zeng"
  - "Cheng Guo"
  - "Yajie Meng"
  - "Xunkun Cheng"
  - "Zhiwei Xu"
  - "Xiangzheng Fu"
  - "Pan Zeng"
  - "Shuting Jin"
  - "Junlin Xu"
doi: "10.24963/ijcai.2026/628"
url: "https://ijcai.org/proceedings/2026/628"
pdf: "https://ijcai.org/proceedings/2026/0628.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "5639-5646"
track: "Main Track"
keywords:
  - "Multidisciplinary Topics and Applications: Bioinformatics"
---

# MultiGeo: Predicting Drug-Target Affinity via Adaptive Multi-Conformation Ensemble Learning

Ruida Zeng, Cheng Guo, Yajie Meng, Xunkun Cheng, Zhiwei Xu, Xiangzheng Fu, Pan Zeng, Shuting Jin, Junlin Xu

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 5639-5646. DOI: [10.24963/ijcai.2026/628](https://doi.org/10.24963/ijcai.2026/628)

## Abstract

Predicting drug–target affinity (DTA) is central to drug discovery, yet most deep learning models rely on a single static protein structure, neglecting the conformational heterogeneity that underlies many binding mechanisms. We propose MultiGeo, a DTA prediction framework that explicitly leverages multiple protein conformations rather than a single snapshot. For each target, MultiGeo starts from an ensemble of structure predictions and uses confidence scores to adaptively select a small set of high-quality, diverse conformers. A hierarchical structural encoder then extracts multi-scale geometric features from these conformers, which are aggregated by a GRU to obtain an ensemble-level representation. To avoid indiscriminately mixing noisy or redundant views, we introduce a disagreement-aware gating mechanism that adaptively fuses this ensemble representation with the dominant structure only when the additional conformers provide complementary information. Finally, ligand–target interactions are modeled via a block-wise cross-attention module that captures multi-perspective dependencies between ligand and protein features. Extensive experiments on multiple DTA benchmarks demonstrate that MultiGeo consistently outperforms state-of-the-art baselines, showing that explicitly encoding conformational diversity yields more accurate and robust affinity prediction.