论文标题
基于相似性的药物组合预测集成的节点嵌入框架
A Node Embedding Framework for Integration of Similarity-based Drug Combination Prediction
论文作者
论文摘要
动机：药物组合是通过提高疗效并降低伴随副作用的明智策略。由于候选化合物之间可能的组合大量组合，详尽的筛选是过于刺激的。目前，大量研究集中在预测潜在的药物组合上。但是，这些方法在性能和可伸缩性上并不完全令人满意。结果：在本文中，我们提出了一个在多重网络（NEMN）中的网络嵌入框架，以预测合成药物组合。基于多重药物相似性网络，我们提供了替代方法来整合来自不同方面的有用信息，并决定每个网络的定量重要性。为了解释NEMN的可行性，我们将我们的框架应用于药物相互作用的数据，在AUPR和ROC方面，它显示出更好的性能。为了进行药物组合预测，我们发现了七个新型药物组合，这些药物组合通过了模型排名最高的预测中的外部来源验证。
Motivation: Drug combination is a sensible strategy for disease treatment by improving the efficacy and reducing concomitant side effects. Due to the large number of possible combinations among candidate compounds, exhaustive screening is prohibitive. Currently, a plenty of studies have focused on predicting potential drug combinations. However, these methods are not entirely satisfactory in performance and scalability. Results: In this paper, we proposed a Network Embedding framework in Multiplex Networks (NEMN) to predict synthetic drug combinations. Based on a multiplex drug similarity network, we offered alternative methods to integrate useful information from different aspects and to decide quantitative importance of each network. To explain the feasibility of NEMN, we applied our framework to the data of drug-drug interactions, on which it showed better performance in terms of AUPR and ROC. For Drug combination prediction, we found seven novel drug combinations which have been validated by external sources among the top-ranked predictions of our model.
