Identifying interactions between known drugs and targets is a major challenge in drug repositioning. In silico prediction of drug-target interaction (DTI) can speed up the expensive and time-consuming experimental work by providing the most potent DTIs. In silico prediction of DTI can also provide insights about the potential drug-drug interaction and promote the exploration of drug side effects. Traditionally, the performance of DTI prediction depends heavily on the descriptors used to represent the drugs and the target proteins. In this paper, to accurately predict new DTIs between approved drugs and targets without separating the targets into different classes, we developed a deep-learning-based algorithmic framework named DeepDTIs. It first abstracts representations from raw input descriptors using unsupervised pretraining and then applies known label pairs of interaction to build a classification model. Compared with other methods, it is found that DeepDTIs reaches or outperforms other state-of-the-art methods. The DeepDTIs can be further used to predict whether a new drug targets to some existing targets or whether a new target interacts with some existing drugs.

Houttuynia cordata (Saururaceae) injection (HCI) is a traditional Chinese medicine used in China. It was chosen as one of eight types of traditional Chinese medicine that play a unique role in severe acute respiratory syndrome (SARS) owing to the effect of curbing inflammation. In order to validate this plausible anti-inflammatory property, the chemical composition of HCI has been analysed by GC/MS, 22 components were identified, and the inflammation induced by carrageenan in the rat pleurisy model and by xylene in the mice ear edema model was adopted to study the anti-inflammatory activity of HCI. Injection of carrageenan into the pleural cavity elicited an acute inflammatory response characterized by protein rich fluid accumulation and leukocyte infiltration in the pleural cavity. The peak inflammatory response was obtained at 24 h when the fluid volume, protein concentration, C-reactive protein and cell infiltration were maximums. The results showed that these parameters were attenuated by HCI at any dose and touched bottom at dose of 0.54 ml/100 g, although less strong than dexamethasone. This drug was also effective in inhibiting xylene induced ear edema, and the percentage of inhibition came to 50% at dose of 80 microl/20 g. The results clearly indicate that HCI have anti-inflammatory activity.

The predicted liquid chromatographic retention times (RTs) of small molecules are not accurate enough for wide adoption in structural identification. In this study, we used the graph neural network to predict the retention time (GNN-RT) from structures of small molecules directly without the requirement of molecular descriptors. The predicted accuracy of GNN-RT was compared with random forests (RFs), Bayesian ridge regression, convolutional neural network (CNN), and a deep-learning regression model (DLM) on a METLIN small molecule retention time (SMRT) dataset. GNN-RT achieved the highest predicting accuracy with a mean relative error of 4.9% and a median relative error of 3.2%. Furthermore, the SMRT-trained GNN-RT model can be transferred to the same type of chromatographic systems easily. The predicted RT is valuable for structural identification in complementary to tandem mass spectra and can be used to assist in the identification of compounds. The results indicate that GNN-RT is a promising method to predict the RT for liquid chromatography and improve the accuracy of structural identification for small molecules.

Kaposi's sarcoma (KS) is an AIDS-defining cancer with aberrant neovascularization caused by KS-associated herpesvirus (KSHV). Although the interaction between HIV-1 and KSHV plays a pivotal role in promoting the aggressive manifestations of KS, the pathogenesis underlying AIDS-KS remains largely unknown. Here we examined HIV-1 Nef protein promotion of KSHV oncoprotein K1-induced angiogenesis. We showed that both internalized and ectopic expression of Nef in endothelial cells synergized with K1 to facilitate vascular tube formation and cell proliferation, and enhance angiogenesis in a chicken CAM model. In vivo experiments further indicated that Nef accelerated K1-induced angiogenesis and tumorigenesis in athymic nu/nu mice. Mechanistic studies revealed that Nef and K1 synergistically activated PI3K/AKT/mTOR signaling by downregulating PTEN. Furthermore, Nef and K1 induced cellular miR-718, which inhibited PTEN expression by directly targeting a seed sequence in the 3′ UTR of its mRNA. Inhibition of miR-718 expression increased PTEN synthesis and suppressed the synergistic effect of Nef- and K1-induced angiogenesis and tumorigenesis. These results indicate that, by targeting PTEN, miR-718 mediates Nef- and K1-induced angiogenesis via activation of AKT/mTOR signaling. Our results demonstrate an essential role of miR-718/AKT/mTOR axis in AIDS-KS and thus may represent an attractive therapeutic target.