Non-coding RNAs (ncRNAs) are a class of RNA molecules that do not have the potential to encode proteins. Meanwhile, they can occupy a significant portion of the human genome and participate in gene expression regulation through various mechanisms. Gestational diabetes mellitus (GDM) is a pathologic condition of carbohydrate intolerance that begins or is first detected during pregnancy, making it one of the most common pregnancy complications. Although the exact pathogenesis of GDM remains unclear, several recent studies have shown that ncRNAs play a crucial regulatory role in GDM. Herein, we present a comprehensive review on the multiple mechanisms of ncRNAs in GDM along with their potential role as biomarkers. In addition, we investigate the contribution of deep learning-based models in discovering disease-specific ncRNA biomarkers and elucidate the underlying mechanisms of ncRNA. This might assist community-wide efforts to obtain insights into the regulatory mechanisms of ncRNAs in disease and guide a novel approach for early diagnosis and treatment of disease.

Medical imaging has been increasingly adopted in the process of medical diagnosis, especially for skin diseases, where diagnoses based on skin pathology are extremely accurate. The diagnostic reports of skin pathology images has the distinguishing features of extreme repetitiveness and rigid formatting. However, reports written by inexperienced radiologists and pathologists can have a high error rate, and even experienced clinicians can find the reporting task both tedious and time-consuming. To address this challenge, this paper studies the automatic generation of diagnostic reports based on images of skin pathologies. A novel deep learning-based image caption framework named the automatic generation network (AGNet), which is an effective network for the automatic generation of skin imaging reports, is proposed. The proposed AGNet consists of four parts: (1) the image model that extracts features and classifies images; (2) the language model that codes data and generates words using comprehensible language; (3) the attention module that connects the \"tail\" of the image model and the \"head\" of the language model, and computes the relationship between images and captions; (4) the embedding and labeling module that processes the input caption data. In case study, The AGNet is verified on a skin pathological image dataset and compared with several state-of-the-art models. The results show that the AGNet achieves the highest scores of the evaluation metrics of image caption among all comparison models, demonstrating the promising performance of the proposed method.

Subtype-selective drugs are of great therapeutic importance as they are expected to be more effective and with less side-effects. However, discovery of subtype selective inhibitors was hampered by the high similarity of the binding sites within subfamilies. In this study, we further evaluated the applicability of \"Three-Dimensional Biologically Relevant Spectrum (BRS-3D)\" for the identification of subtype-selective inhibitors. A case study was performed on monoamine oxidase, which has two subtypes related to distinct diseases. The inhibitory activity against MAO-A/B of 347 compounds experimentally tested in this research was reported. Compound M124 (5H-thiazolo[3,2-a]pyrimidin-5-one) with IC50 less than 100 nM (SI = 23) was selected as a probe to investigate the structure selectivity relationship. Similarity search led to the identification of compound M229 and M249 with IC50 values of 7.4 nM, 4 nM and acceptable selectivity index over MAO-A (M229 SI > 1351, M249 SI > 2500). The molecular basis for subtype selectivity was explored through docking study and attention based DNN model. Additionally, in silico ADME properties were characterized. Accordingly, it is found that BRS-3D is a robust method for subtype selectivity in the early stage of drug discovery and the compounds reported here can be promising leads for further experimental analysis.