Learning long- and short-term dependencies for improving drug-target binding affinity prediction using transformer and edge contraction pooling.

The accurate identification of drug-target affinity (DTA) is crucial for advancements in drug discovery and development. Many deep learning-based approaches have been devised to predict drug-target binding affinity accurately, exhibiting notable improvements in performance. However, the existing prediction methods often fall short of capturing the global features of proteins. In this study, we proposed a novel model called ETransDTA, specifically designed for predicting drug-target binding affinity. ETransDTA combines convolutional layers and transformer, allowing for the simultaneous extraction of both global and local features of target proteins. Additionally, we have integrated a new graph pooling mechanism into the topology adaptive graph convolutional network (TAGCN) to enhance its capacity for learning feature representations of chemical compounds. The proposed ETransDTA model has been evaluated using the Davis and Kinase Inhibitor BioActivity (KIBA) datasets, consistently outperforming other baseline methods. The evaluation results on the KIBA dataset reveal that our model achieves the lowest mean square error (MSE) of 0.125, representing a 0.6% reduction compared to the lowest-performing baseline method. Furthermore, the incorporation of queries, keys and values produced by the stacked convolutional neural network (CNN) enables our model to better integrate the local and global context of protein representation, leading to further improvements in the accuracy of DTA prediction.

Increased microRNA-93-5p inhibits osteogenic differentiation by targeting bone morphogenetic protein-2.

BACKGROUND AND PURPOSE: Trauma-induced osteonecrosis of the femoral head (TIONFH) is a major complication of femoral neck fractures. Degeneration and necrosis of subchondral bone can cause collapse, which results in hip joint dysfunction in patients. The destruction of bone metabolism homeostasis is an important factor for osteonecrosis. MicroRNAs (miRNAs) have an important role in regulating osteogenic differentiation, but the mechanisms underlying abnormal bone metabolism of TIONFH are poorly understood. In this study, we screened specific miRNAs in TIONFH by microarray and further explored the mechanism of osteogenic differentiation. DESIGN: Blood samples from patients with TIONFH and patients without necrosis after trauma were compared by microarray, and bone collapse of necrotic bone tissue was evaluated by micro-CT and immunohistochemistry. To confirm the relationship between miRNA and osteogenic differentiation, we conducted cell culture experiments. We found that many miRNAs were significantly different, including miR-93-5p; the increase in this miRNA was verified by Q-PCR. Comparison of the tissue samples showed that miR-93-5p expression increased, and alkaline phosphatase (ALP) and osteopontin (OPN) levels decreased, suggesting miR-93-5p may be involved in osteogenic differentiation. Further bioinformatics analysis indicated that miR-93-5p can target bone morphogenetic protein 2 (BMP-2). A luciferase gene reporter assay was performed to confirm these findings. By simulating and/or inhibiting miR-93-5p expression in human bone marrow mesenchymal stem cells, we confirmed that osteogenic differentiation-related indictors, including BMP-2, Osterix, Runt-related transcription factor, ALP and OPN, were decreased by miR-93-5p. CONCLUSION: Our study showed that increased miR-93-5p in TIONFH patients inhibited osteogenic differentiation, which may be associated with BMP-2 reduction. Therefore, miR-93-5p may be a potential target for prevention of TIONFH.

LMP1 stimulates the transcription of eIF4E to promote the proliferation, migration and invasion of human nasopharyngeal carcinoma.

Eukaryotic translation initiation factor 4E (eIF4E) is the rate-limiting translation initiation factor for many oncogenes. Previous studies have shown eIF4E overexpression in nasopharyngeal carcinoma (NPC). We aimed to study whether viral oncogene latent membrane protein 1 (LMP1) stimulates the transcription of eIF4E to promote NPC malignancy. In NPC cell lines (CNE1 and CNE2), ectopic LMP1 significantly increased the mRNA and protein levels of eIF4E and the transcriptional activity of the eIF4E promoter in a LMP1-plasmid-transfected dose-dependent manner. As a backward experiment, knocking down of LMP1 significantly reduced eIF4E mRNA in B95-8 cells. In the high LMP1 expression condition, knocking down of c-Myc significantly reduced eIF4E mRNA in both NPC and B95-8 cells, and knocking down of eIF4E significantly inhibited the tumor proliferation, migration and invasion promoted by LMP1. The results indicated that LMP1 stimulates the transcription of eIF4E via c-Myc to promote NPC. To the best of our knowledge, this is the first evidence that LMP1 stimulates the transcription of eIF4E. This might be an important cause of the overexpression of eIF4E in NPC and be the novel mechanism by which LMP1 initiates cancer. LMP1-stimulated eIF4E initiates the translation of those oncogenes transcriptionally activated by LMP1 to amplify and pass down the carcinogenesis signals launched by LMP1.

HPV E6 induces eIF4E transcription to promote the proliferation and migration of cervical cancer.

Increasing evidence has placed eukaryotic translation initiation factor 4E (eIF4E) at the hub of tumor development and progression. Several studies have reported that eIF4E is over-expressed in cervical cancer; however, the mechanism remains elusive. The results of this study further confirm over-expression of eIF4E in cervical cancer tumors and cell lines, and we have discovered that the transcription of eIF4E is induced by protein E6 of the human papillomavirus (HPV). Moreover, regulation of eIF4E by E6 significantly influences cell proliferation, the cell cycle, migration, and apoptosis. Therefore, eIF4E emerges as a key player in tumor development and progression and a potential target for CC treatment and prevention.