A miRNA-disease association prediction model based on tree-path global feature extraction and fully connected artificial neural network with multi-head self-attention mechanism.

BACKGROUND: MicroRNAs (miRNAs) emerge in various organisms, ranging from viruses to humans, and play crucial regulatory roles within cells, participating in a variety of biological processes. In numerous prediction methods for miRNA-disease associations, the issue of over-dependence on both similarity measurement data and the association matrix still hasn't been improved. In this paper, a miRNA-Disease association prediction model (called TP-MDA) based on tree path global feature extraction and fully connected artificial neural network (FANN) with multi-head self-attention mechanism is proposed. The TP-MDA model utilizes an association tree structure to represent the data relationships, multi-head self-attention mechanism for extracting feature vectors, and fully connected artificial neural network with 5-fold cross-validation for model training. RESULTS: The experimental results indicate that the TP-MDA model outperforms the other comparative models, AUC is 0.9714. In the case studies of miRNAs associated with colorectal cancer and lung cancer, among the top 15 miRNAs predicted by the model, 12 in colorectal cancer and 15 in lung cancer were validated respectively, the accuracy is as high as 0.9227. CONCLUSIONS: The model proposed in this paper can accurately predict the miRNA-disease association, and can serve as a valuable reference for data mining and association prediction in the fields of life sciences, biology, and disease genetics, among others.

Effects of Parathyroidectomy on Plasma iPTH and (1-84) PTH Levels in Patients with Stage 5 Chronic Kidney Disease.

Currently, the second-generation intact parathyroid hormone (iPTH) assay is commonly used for measuring PTH levels. The iPTH assay detects both full-length (1-84)PTH and (7-84)PTH fragments, which have antagonistic effects on (1-84)PTH in bones and kidneys. The third-generation PTH assay is specific for (1-84)PTH. This study examined the features of different PTH fragments in stage 5 chronic kidney disease (CKD) and the effects of parathyroidectomy (PTX) on the above markers in severe secondary hyperparathyroidism (SHPT) patients. The cross-sectional study included 262 stage 5 CKD patients and 90 controls. A prospective follow-up study was then conducted in 34 PTX patients. Second- and third-generation assays were used to measure plasma iPTH and (1-84)PTH levels, respectively. Circulating (7-84)PTH levels were calculated by subtracting the (1-84)PTH value from the iPTH value. Different plasma PTH fragments were higher, and (1-84)PTH/iPTH was lower in CKD patients than in controls. Plasma (1-84)PTH and (7-84)PTH concentrations increased as iPTH levels increased, and (7-84)PTH increased more evidently. Plasma iPTH, (1-84)PTH and (7-84)PTH levels were 1530.5 (885.0-2111.5) pg/ml, 683.1 (431.4-1018.0) pg/ml, and 739.3 (452.6-1261.0) pg/ml, respectively, in PTX patients. Plasma iPTH, (1-84)PTH and (7-84)PTH concentrations decreased considerably, and the (1-84)PTH/iPTH ratio increased after PTX (median follow-up interval: 10.9 months). Stage 5 CKD patients had higher plasma levels of different PTH fragments, and lower (1-84)PTH/iPTH ratio. PTX could significantly reverse these abnormalities in severe SHPT patients. The iPTH assay overestimated the function of the parathyroid glands; thus, the third-generation PTH assay is likely better for the management of CKD patients.

A lncRNA-disease association prediction model based on the two-step PU learning and fully connected neural networks.

Long non-coding RNAs (lncRNAs) have been shown to play a regulatory role in various processes of human diseases. However, lncRNA experiments are inefficient, time-consuming and highly subjective, so that the number of experimentally verified associations between lncRNA and diseases is limited. In the era of big data, numerous machine learning methods have been proposed to predict the potential association between lncRNA and diseases, but the characteristics of the associated data were seldom explored. In these methods, negative samples are randomly selected for model training and the model is prone to learn the potential positive association error, thus affecting the prediction accuracy. In this paper, we proposed a cyclic optimization model of predicting lncRNA-disease associations (COPTLDA in short). In COPTLDA, the two-step training strategy is adopted to search for the samples with the greater probability of being negative examples from unlabeled samples and the determined samples are treated as negative samples, which are combined together with known positive samples to train the model. The searching and training steps are repeated until the best model is obtained as the final prediction model. In order to evaluate the performance of the model, 30% of the known positive samples are used to calculate the model accuracy and 10% of positive samples are used to calculate the recall rate of the model. The sampling strategy used in this paper can improve the accuracy and the AUC value reaches 0.9348. The results of case studies showed that the model could predict the potential associations between lncRNA and malignant tumors such as colorectal cancer, gastric cancer, and breast cancer. The predicted top 20 associated lncRNAs included 10 colorectal cancer lncRNAs, 2 gastric cancer lncRNAs, and 8 breast cancer lncRNAs.

Regulatory B Cells and Its Role in Central Nervous System Inflammatory Demyelinating Diseases.

Regulatory B (Breg) cells represent a population of suppressor B cells that participate in immunomodulatory processes and inhibition of excessive inflammation. The regulatory function of Breg cells have been demonstrated in mice and human with inflammatory diseases, cancer, after transplantation, and particularly in autoinflammatory disorders. In order to suppress inflammation, Breg cells produce anti-inflammatory mediators, induce death ligand-mediated apoptosis, and regulate many kinds of immune cells such as suppressing the proliferation and differentiation of effector T cell and increasing the number of regulatory T cells. Central nervous system Inflammatory demyelinating diseases (CNS IDDs) are a heterogeneous group of disorders, which occur against the background of an acute or chronic inflammatory process. With the advent of monoclonal antibodies directed against B cells, breakthroughs have been made in the treatment of CNS IDDs. Therefore, the number and function of B cells in IDDs have attracted attention. Meanwhile, increasing number of studies have confirmed that Breg cells play a role in alleviating autoimmune diseases, and treatment with Breg cells has also been proposed as a new therapeutic direction. In this review, we focus on the understanding of the development and function of Breg cells and on the diversification of Breg cells in CNS IDDs.

Robust and Sparse Linear Discriminant Analysis via an Alternating Direction Method of Multipliers.

In this paper, we propose a robust linear discriminant analysis (RLDA) through Bhattacharyya error bound optimization. RLDA considers a nonconvex problem with the L1 -norm operation that makes it less sensitive to outliers and noise than the L2 -norm linear discriminant analysis (LDA). In addition, we extend our RLDA to a sparse model (RSLDA). Both RLDA and RSLDA can extract unbounded numbers of features and avoid the small sample size (SSS) problem, and an alternating direction method of multipliers (ADMM) is used to cope with the nonconvexity in the proposed formulations. Compared with the traditional LDA, our RLDA and RSLDA are more robust to outliers and noise, and RSLDA can obtain sparse discriminant directions. These findings are supported by experiments on artificial data sets as well as human face databases.

Comparative Investigation of 0.5Li2MnO3·0.5LiNi0.5Co0.2Mn0.3O2 Cathode Materials Synthesized by Using Different Lithium Sources.

Lithium-rich manganese-based cathode materials has been attracted enormous interests as one of the most promising candidates of cathode materials for next-generation lithium ion batteries because of its high theoretic capacity and low cost. In this study, 0.5Li2MnO3·0.5LiNi0.5Co0.2Mn0.3O2 materials are synthesized through a solid-state reaction by using different lithium sources, and the synthesis process and the reaction mechanism are investigated in detail. The morphology, structure, and electrochemical performances of the material synthesized by using LiOH·H2O, Li2CO3, and CH3COOLi·2H2O have been analyzed by using Thermo gravimetric analysis (TGA), X-ray diffraction (XRD), Scanning electron microscope (SEM), Transmission electron microscope (TEM), X-ray photoelectron spectroscopy (XPS), and electrochemical measurements. The 0.5Li2MnO3·0.5LiNi0.5Co0.2Mn0.3O2 material prepared by using LiOH·H2O displays uniform morphology with nano particle and stable layer structure so that it suppresses the first cycle irreversible reaction and structure transfer, and it delivers the best electrochemical performance. The results indicate that LiOH·H2O is the best choice for the synthesis of the 0.5Li2MnO3·0.5LiNi0.5Co0.2Mn0.3O2 material.

Robust recursive absolute value inequalities discriminant analysis with sparseness.

In this paper, we propose a novel absolute value inequalities discriminant analysis (AVIDA) criterion for supervised dimensionality reduction. Compared with the conventional linear discriminant analysis (LDA), the main characteristics of our AVIDA are robustness and sparseness. By reformulating the generalized eigenvalue problem in LDA to a related SVM-type "concave-convex" problem based on absolute value inequalities loss, our AVIDA is not only more robust to outliers and noises, but also avoids the SSS problem. Moreover, the additional L1-norm regularization term in the objective makes sure sparse discriminant vectors are obtained. A successive linear algorithm is employed to solve the proposed optimization problem, where a series of linear programs are solved. The superiority of our AVIDA is supported by experimental results on artificial examples as well as benchmark image databases.