HLA antigen presentation prediction method and system based on multi-modal depth coding

The invention discloses an HLA antigen presentation prediction method based on multi-modal depth coding. The method comprises the following steps: 1) coding known sequence information by adopting multiple different deep neural networks; 2) introducing existing literatures and calculation tools to calculate an affinity index; and 3) fusing multi-modal features to obtain a prediction score and performing prediction. Different from a previous method only based on biological experiment or affinity index prediction, the system scheme can efficiently fuse multi-modal information, and more accurate and efficient prediction is carried out. And moreover, the method has flexible expansibility on modules for processing negative samples of different methods and processing data imbalance, and can better adapt to a real drug research and development production environment. 本发明公开了一种基于多模态深度编码的HLA抗原呈递预测方法，包括：1)采用多种不同的深度神经网络来编码已知序列信息2)引入已有的文献与计算工具计算亲和力指数3)多模态特征融合得到预测分数并进行预测。不同于以往的仅基于生物实验或亲和力指数预测的方法，本系统方案能高效地融合多模态信息，进行更加准确高效的预测