18 April 2024 in IEEE/ACM Transactions on Computational Biology and Bioinformatics
As medical safety and drug regulation gain heightened attention, the detection of spurious drug-drug interactions (DDI) has become key in healthcare. Although current research using graph neural networks (GNNs) to predict DDI has shown impressive results, it often fails to account for false DDI in the constructed DDI networks. Such inaccuracies caused by data errors, false alarms, or incorrect drug details can skew the network's structure and hinder the accuracy of GNN-based predictions. To tackle this challenge, we propose ANSM, a network-enhancement method specifically designed to identify and attenuate spurious links between drugs for ensuring the accuracy of DDI networks. ANSM integrates three key components: the feature extractor, the network optimizer, and the discriminative classifier. The feature extractor captures local structural features from drug node pairs, while the network optimizer leverages network information to improve feature extraction and reduce the impact of spurious DDI links. The discriminative classifier then identifies potential spurious links. Experimental results demonstrate that ANSM outperforms state-of-the-art methods in identifying spurious DDI.
14 June 2023 in IEEE/ACM Transactions on Computational Biology and Bioinformatics
With the growing popularity of artificial intelligence in drug discovery, many deep-learning technologies have been used to automatically predict unknown drug-target interactions (DTIs). A unique challenge in using these technologies to predict DTIs is fully exploiting the knowledge diversity across different interaction types, such as drug-enzyme, drug-target, drug-path, and drug-structure types. Unfortunately, existing methods tend to learn the specifical knowledge on each interaction type and they usually ignore the knowledge diversity across different interaction types. Therefore, we propose a multitype perception method (MPM) for DTI prediction by exploiting knowledge diversity across different link types. The method consists of a type perceptor and a multitype predictor. The type perceptor learns distinguished edge representations by retaining the specifical features across different interaction types; this maximizes the prediction performance for each interaction type. The multitype predictor evaluates the type similarity between the type perceptor and potential interactions, and a domain gate module is further reconstructed to assign an adaptive weight to each type perceptor. Based on the type preceptor and the multitype predictor, our proposed MPM is proposed to exploit the knowledge diversity across different interaction types for DTI prediction. Extensive experiments demonstrate that our proposed MPM outperforms the state-of-the-art methods in DTI prediction.
12 June 2026 in Expert Systems with Applications