Institute of Intelligent Emergency Information Processing, Institute of Disaster Prevention, Langfang 065201, China.
School of Emergency Management, Institute of Disaster Prevention, Langfang 065201, China.
Comput Math Methods Med. 2022 Jun 9;2022:8693746. doi: 10.1155/2022/8693746. eCollection 2022.
Drug combinations have recently been studied intensively due to their critical role in cancer treatment. Computational prediction of drug synergy has become a popular alternative strategy to experimental methods for anticancer drug synergy predictions. In this paper, a deep learning model called DCE-DForest is proposed to predict the synergistic effect of drug combinations. To sufficiently extract drug information, the paper leverages BERT (Bidirectional Encoder Representations from Transformers) to encode the drug and the deep forest to model the nonlinear relationship between the drugs and cell lines. The experimental results on the synergy datasets demonstrate that the proposed method consistently shows superior performance over the other machine learning models.
由于在癌症治疗中的关键作用，药物组合最近受到了广泛关注。计算预测药物协同作用已经成为预测抗癌药物协同作用的实验方法的一种替代策略。在本文中，提出了一种称为 DCE-DForest 的深度学习模型，用于预测药物组合的协同效应。为了充分提取药物信息，本文利用 BERT（来自 Transformer 的双向编码器表示）对药物进行编码，并用深度森林对药物和细胞系之间的非线性关系进行建模。协同作用数据集上的实验结果表明，所提出的方法在性能上明显优于其他机器学习模型。
