College of Computer Science and Technology, China University of Petroleum, Qingdao 266580, China; Department of Artificial Intelligence, Faculty of Computer Science, Polytechnical University of Madrid, Campus de Montegancedo, Boadilla del Monte 28660, Madrid, Spain.
College of Computer Science and Technology, China University of Petroleum, Qingdao 266580, China.
Methods. 2022 Aug;204:269-277. doi: 10.1016/j.ymeth.2022.02.007. Epub 2022 Feb 24.
Predicting drug-target interactions (DTIs) is essential for both drug discovery and drug repositioning. Recently, deep learning methods have achieved relatively significant performance in predicting DTIs. Generally, it needs a large amount of approved data of DTIs to train the model, which is actually tedious to obtain. In this work, we propose DeepFusion, a deep learning based multi-scale feature fusion method for predicting DTIs. To be specific, we generate global structural similarity feature based on similarity theory, convolutional neural network and generate local chemical sub-structure semantic feature using transformer network respectively for both drug and protein. Data experiments are conducted on four sub-datasets of BIOSNAP, which are 100%, 70%, 50% and 30% of BIOSNAP dataset. Particularly, using 70% sub-dataset, DeepFusion achieves ROC-AUC and PR-AUC by 0.877 and 0.888, which is close to the performance of some baseline methods trained by the whole dataset. In case study, DeepFusion achieves promising prediction results on predicting potential DTIs in case study.
预测药物-靶标相互作用（DTIs）对于药物发现和药物重定位都至关重要。最近，深度学习方法在预测 DTIs 方面取得了相对显著的性能。通常，它需要大量已批准的 DTIs 数据来训练模型，而实际上这是一项繁琐的工作。在这项工作中，我们提出了 DeepFusion，这是一种基于深度学习的多尺度特征融合方法，用于预测 DTIs。具体来说，我们基于相似理论、卷积神经网络生成全局结构相似性特征，以及使用转换器网络生成药物和蛋白质的局部化学子结构语义特征。我们在 BIOSNAP 的四个子数据集上进行了数据实验，这四个子数据集分别为 BIOSNAP 数据集的 100%、70%、50%和 30%。特别是，在使用 70%的子数据集时，DeepFusion 在 ROC-AUC 和 PR-AUC 上的得分分别为 0.877 和 0.888，接近使用整个数据集训练的一些基线方法的性能。在案例研究中，DeepFusion 在案例研究中预测潜在 DTIs 方面取得了有前途的预测结果。
