Deep Bi-LSTM Networks for Sequential Recommendation.

Recent years have seen a surge in approaches that combine deep learning and recommendation systems to capture user preference or item interaction evolution over time. However, the most related work only consider the sequential similarity between the items and neglects the item content feature information and the impact difference of interacted items on the next items. This paper introduces the deep bidirectional long short-term memory (LSTM) and self-attention mechanism into the sequential recommender while fusing the information of item sequences and contents. Specifically, we deal with the issues in a three-pronged attack: the improved item embedding, weight update, and the deep bidirectional LSTM preference learning. First, the user-item sequences are embedded into a low-dimensional item vector space representation via Item2vec, and the class label vectors are concatenated for each embedded item vector. Second, the embedded item vectors learn different impact weights of each item to achieve item awareness via self-attention mechanism; the embedded item vectors and corresponding weights are then fed into the bidirectional LSTM model to learn the user preference vectors. Finally, the top similar items in the preference vector space are evaluated to generate the recommendation list for users. By conducting comprehensive experiments, we demonstrate that our model outperforms the traditional recommendation algorithms on Recall@20 and Mean Reciprocal Rank (MRR@20).

EBV-LMP1-targeted DNAzyme induces DNA damage and causes cell cycle arrest in LMP1-positive nasopharyngeal carcinoma cells.

This study aimed to determine the molecular mechanisms underlying the effect of the LMP1-targeted DNAzyme 1 (DZ1) on cell cycle progression in nasopharyngeal carcinoma (NPC) cells. We showed that the active DZ1 inhibited the expression of latent membrane protein 1 (LMP1) and induced a G1 phase arrest. In addition, this cell cycle deregulation was shown to be accompanied by upregulation of the DNA damage marker γ-H2AX, downregulation of the DNA damage response factor p-p53-Ser15 and cell proliferation inhibition. To investigate what affected the cell cycle progression, we examined the expression of two checkpoint-related cyclins and cyclin-dependent kinases (CDKs). We found a decrease of cyclin D1 and cyclin E protein levels at 24 h from the DZ1 treatment. Moreover, we observed inhibition of CDK4 activity and decreased cyclin D1 expression in the complexes immunoprecipitated with CDK4 antibody. We also found a reduction in cdc2 phosphorylation at Thr161 which partially stands for the cdc2 kinase activity in DZ1-treated CNE1-LMP1 cells, although the downregulation of LMP1 expression had no effect on the cyclin B1 and cdc2 expression. Further, we analyzed changes in cdc2 kinase activity induced by DZ1 and found that the downregulation of the LMP1 expression resulted in a 5-fold reduction in cdc2 kinase activity in CNE1-LMP1. The data suggest that the downregulation of the LMP1 expression by DZ1 was able to induce DNA damage, which then further inhibited the cell proliferation and resulted in malfunction of cell cycle checkpoints that led to G1 phase arrest and the decrease in number of cells in G2/M phase.