arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2512.22007cs.LG
收藏

DuaDeep-SeqAffinity:双流深度学习框架用于仅序列抗原-抗体亲和力预测

DuaDeep-SeqAffinity: Dual-Branch Deep Learning for Tri-Stream Sequence-Based Antibody--Antigen Affinity Prediction

National School of Artificial Intelligence (ENSIA)(人工智能国家学校)

Aicha Boutorh, Soumia Bouyahiaoui, Manel Kara Laouar, Sara Belhadj, Nour El Yakine Guendouz, Asma Boutorh

AI总结:

DuaDeep-SeqAffinity通过双流深度学习框架,仅利用氨基酸序列预测抗原-抗体亲和力,优于现有方法并提高预测精度。

摘要:

预测抗原与抗体之间的结合亲和力对于药物发现和疫苗开发至关重要。传统计算方法往往依赖于实验确定的三维结构,这些结构稀缺且获取成本高。本文介绍了DuaDeep-SeqAffinity,一种新的仅序列深度学习框架,该框架通过双流混合架构仅使用氨基酸序列预测亲和力分数。我们的方法利用预训练的ESM-2蛋白质语言模型嵌入,结合一维卷积神经网络(CNNs)用于局部motif检测和Transformer编码器用于全局上下文表示。随后的融合模块整合这些多方面的特征,然后传递给全连接网络进行最终分数回归。实验结果表明,DuaDeep-SeqAffinity显著优于单独的架构组件和现有的最先进(SOTA)方法。DuaDeep实现了优于0.688的皮尔逊相关系数,0.460的R平方值和0.737的均方根误差,超过了单一分支变体ESM-CNN和ESM-Transformer。值得注意的是,该模型实现了曲线下面积(AUC)为0.890,超过了仅序列基准,甚至超过了结构-序列混合模型。这些发现证明了高保真的序列嵌入可以捕捉到通常保留给结构建模的必要结合模式。通过消除对三维结构的依赖,DuaDeep-SeqAffinity为大规模序列库的高通量筛选提供了高度可扩展和高效的方法,显著加速了治疗发现流程。

英文摘要:

DuaDeep-SeqAffinity is a sequence-only deep learning framework that predicts antibody--antigen binding affinity directly from primary amino acid sequences, avoiding the cost and scarcity of resolved three-dimensional structures. The antigen and the antibody heavy and light chains are processed as three independent streams, each embedded with a frozen ESM-2 protein language model and passed through parallel Transformer and convolutional neural network (CNN) branches before late fusion, a decoupled design intended to preserve local complementarity-determining region (CDR) signal that monolithic encoders can dilute. On a sequence-disjoint split of the AbRank benchmark, the model achieves a Pearson correlation of 0.683, an R^2 of 0.460, and a pairwise ranking AUC of 0.895, significantly outperforming single-branch ablations (paired t-test, p < 0.05). Attention-map and gradient-based saliency analyses further show that the model preferentially attends to CDR loops and candidate epitope residues, supporting its use as a scalable, structure-free tool for high-throughput antibody screening.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑