# CAME-AB: Cross-Modality Attention with Mixture-of-Experts for Antibody Binding Site Prediction

arXiv: https://arxiv.org/abs/2509.06465 · HTML: https://mlsanity.com/paper/2509.06465/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Medical and biomedical MoE, one of 12 directions in Mixture-of-experts.
- Builds on 7 of its 37 references that are in the map.
- First posted on arXiv 8 Sep 2025.

## Placed in

- Mixture-of-experts / [Medical and biomedical MoE](https://mlsanity.com/moe/medical-and-biomedical-moe/)

## Builds on (most cited first, 7 of 7)

- [Semi-Supervised Classification with Graph Convolutional Networks](https://mlsanity.com/paper/1609.02907/) (2016)
- [Outrageously Large Neural Networks: The Sparsely-Gated Mixture-of-Experts Layer](https://mlsanity.com/paper/1701.06538/) (2017)
- [Supervised Contrastive Learning](https://mlsanity.com/paper/2004.11362/) (2020)
- [On Layer Normalization in the Transformer Architecture](https://mlsanity.com/paper/2002.04745/) (2020)
- [Diversify and Match: A Domain Adaptive Representation Learning Paradigm for Object Detection](https://mlsanity.com/paper/1905.05396/) (2019)
- [SWA Object Detection](https://mlsanity.com/paper/2012.12645/) (2020)
- [SemiGNN-PPI: Self-Ensembling Multi-Graph Neural Network for Efficient and Generalizable Protein-Protein Interaction Prediction](https://mlsanity.com/paper/2305.08316/) (2023)

## Abstract (arXiv)

Antibody binding site prediction plays a pivotal role in computational immunology and therapeutic antibody design. Existing sequence or structure methods rely on single-view features and fail to identify antibody-specific binding sites on the antigens. In this paper, we propose CAME-AB, a novel Cross-modality Attention framework with a Mixture-of-Experts (MoE) backbone for robust antibody binding site prediction. CAME-AB integrates five biologically grounded modalities, including raw amino acid encodings, BLOSUM substitution profiles, pretrained language model embeddings, structure-aware features, and GCN-refined biochemical graphs, into a unified multimodal representation. To enhance adaptive cross-modal reasoning, we propose an adaptive modality fusion module that learns to dynamically weight each modality based on its global relevance and input-specific contribution. A Transformer encoder combined with an MoE module further promotes feature specialization and capacity expansion. We additionally incorporate a supervised contrastive learning objective to explicitly shape the latent space geometry, encouraging intra-class compactness and inter-class separability. To improve optimization stability and generalization, we apply stochastic weight averaging during training. Extensive experiments on benchmark antibody-antigen datasets demonstrate that CAME-AB consistently outperforms strong baselines on multiple metrics, including Precision, Recall, F1-score, AUC-ROC, and MCC. Ablation studies further validate the effectiveness of each architectural component and the benefit of multimodal feature integration. The model implementation details and the codes are available on https://anonymous.4open.science/r/CAME-AB-C525