29 December 2020 in IEEE Journal of Biomedical and Health Informatics
Adverse drug-drug interaction (ADDI) becomes a significant threat to public health. Despite the detection of ADDIs is experimentally implemented in the early development phase of drug design, many potential ADDIs are still clinically explored by accidents, leading to a large number of morbidity and mortality. Several computational models are designed for ADDI prediction. However, they take no consideration of drug dependency, although many drugs usually produce synergistic effects and own highly mutual dependency in treatments, which contains underlying information about ADDIs and benefits ADDI prediction. In this paper, we design a dependent network to model the drug dependency and propose an attribute supervised learning model Probabilistic Dependent Matrix Tri-Factorization (PDMTF) for ADDI prediction. In particular, PDMTF incorporates two drug attributes, molecular structure and side effect, and their correlation to model the adverse interactions among drugs. The dependent network is represented by a dependent matrix, which is first formulated by the row precision matrix of the predicted attribute matrices and then regularized by the molecular structure similarities among drugs. Meanwhile, an efficient alternating algorithm is designed for solving the optimization problem of PDMTF. Experiments demonstrate the superior performance of the proposed model when compared with eight baselines and its two variants.
16 December 2021 in IEEE Transactions on Pattern Analysis and Machine Intelligence
Adverse drug-drug interaction (ADDI) is a significant life-threatening issue, posing a leading cause of hospitalizations and deaths in healthcare systems. This paper proposes a unified Multi-Attribute Discriminative Representation Learning (MADRL) model for ADDI prediction. Unlike the existing works that equally treat features of each attribute without discrimination and do not consider the underlying relationship among drugs, we first develop a regularized optimization problem based on CUR matrix decomposition for joint representative drug and discriminative feature selection such that the selected drugs and features can well approximate the original feature spaces and the critical factors discriminative to ADDIs can be properly explored. Different from the existing models that ignore the consistent and unique properties among attributes, a Generative Adversarial Network (GAN) framework is then designed to capture the inter-attribute shared and intra-attribute specific representations of adverse drug pairs for exploiting their consensus and complementary information in ADDI prediction. Meanwhile, MADRL is compatible with any kind of attributes and capable of exploring their respective effects on ADDI prediction. An iterative algorithm based on the alternating direction method of multipliers is developed for optimization. Experiments on publicly available dataset demonstrate the effectiveness of MADRL when compared with eleven baselines and its six variants.
19 November 2022 in Expert Systems with Applications
2 March 2020 in IEEE Access
Community is an important topological characteristic of complex networks, which is significant for understanding the structural feature and organizational function of networks, and community detection has recently attracted considerable research effort. Among community detection methods, label propagation technology is widely used because of its linear time complexity. However, due to the randomness of the node order of label updating and the order of label launching in label propagation, the instability of community detection approaches based on label propagation becomes a challenge. In this paper, a new label propagation algorithm, Node Ability based Label Propagation Algorithm (NALPA), is proposed to discover communities in networks. Inspired from human society and radar transmission, we design four node abilities (propagation ability, attraction ability, launch ability and acceptance ability), label influence and a new label propagation mechanism to deal with the instability and enhance the efficiency. Experimental results on 42 synthetic and 14 real-world networks demonstrate that NALPA outperforms state-of-the-art approaches in most cases. In a case study, NALPA is applied to a drug network in Traditional Chinese Medicine (TCM) and can discover the drug communities where drugs have similar efficacy for treating Chronic GlomeruloNephritis (CGN).
16 April 2021 in IEEE/ACM Transactions on Audio, Speech, and Language Processing
Chinese word embedding models capture Chinese semantics based on the character feature of Chinese words and the internal features of Chinese characters such as radical, component, stroke, structure and pinyin. However, some features are overlapping and most methods do not consider their relevance. Meanwhile, they express words as point vectors that cannot better capture different aspect semantics of Chinese words. In this paper, we propose a Feature Subsequence based Probability Representation Model (FSPRM) for learning Chinese word embeddings, in which we first integrate the morphological and phonetic features (stroke, structure and pinyin) of Chinese characters and learn their relevance by designing a feature subsequence to capture relatively comprehensive semantics of Chinese words, then feature probability distribution is proposed for capturing different aspect meanings of Chinese words based on the three internal features and probability representation by estimating its mean as the sum of feature subsequences. Chinese words with similar features may have similar semantics, then we map Chinese words to feature probability distributions and design a similarity-based objective for predicting the contextual words of the target word to learn their semantics. Extensive experiments on word analogy, word similarity, text classification and named entity recognition tasks demonstrate that the proposed method outperforms most state-of-the-art approaches.
3 March 2020 in IEEE Transactions on Knowledge and Data Engineering
Adverse drug-drug interaction (ADDI) is referred to as a situation where the unpleasant or adverse effects caused by the co-administration of two drugs, which becomes a significant problem for public health. With the increasing availability of healthcare data, many methods are proposed for ADDI prediction. However, these methods usually work in a "nondiscriminatory" manner, i.e., they treat each feature without discrimination and equally incorporate all features into the predictive models. In practice, only a few features are essentially discriminative and relevant to ADDIs. In this paper, we propose a Dependence Guided Discriminative Feature Selection (DGDFS) model for ADDI prediction. In DGDFS, two drug attributes, molecular structure and side effect are adopted to model the adverse interaction among drugs and $l_{2,0}$ -norm equality constraints are introduced to select discriminative molecular substructures and side effects for ADDI prediction. Besides, three dependence guided terms, i.e., the dependence between molecular structure and ADDI, the dependence between side effect and ADDI, and the dependence between molecular structure and side effect, are designed to guide feature selection. An iterative algorithm based on the alternating direction method of multipliers is developed for optimization. Experimental results indicate the effectiveness of DGDFS compared with fourteen baselines and its three variants.