Prediction of miRNA-disease associations based on strengthened hypergraph convolutional autoencoder.

Most existing graph neural network-based methods for predicting miRNA-disease associations rely on initial association matrices to pass messages, but the sparsity of these matrices greatly limits performance. To address this issue and predict potential associations between miRNAs and diseases, we propose a method called strengthened hypergraph convolutional autoencoder (SHGAE). SHGAE leverages multiple layers of strengthened hypergraph neural networks (SHGNN) to obtain robust node embeddings. Within SHGNN, we design a strengthened hypergraph convolutional network module (SHGCN) that enhances original graph associations and reduces matrix sparsity. Additionally, SHGCN expands node receptive fields by utilizing hyperedge features as intermediaries to obtain high-order neighbor embeddings. To improve performance, we also incorporate attention-based fusion of self-embeddings and SHGCN embeddings. SHGAE predicts potential miRNA-disease associations using a multilayer perceptron as the decoder. Across multiple metrics, SHGAE outperforms other state-of-the-art methods in five-fold cross-validation. Furthermore, we evaluate SHGAE on colon and lung neoplasms cases to demonstrate its ability to predict potential associations. Notably, SHGAE also performs well in the analysis of gastric neoplasms without miRNA associations.

LUNCRW: Prediction of potential lncRNA-disease associations based on unbalanced neighborhood constraint random walk.

Accumulating evidence suggests that long non-coding RNAs (lncRNAs) are associated with various complex human diseases. They can serve as disease biomarkers and hold considerable promise for the prevention and treatment of various diseases. The traditional random walk algorithms generally exclude the effect of non-neighboring nodes on random walking. In order to overcome the issue, the neighborhood constraint (NC) approach is proposed in this study for regulating the direction of the random walk by computing the effects of both neighboring nodes and non-neighboring nodes. Then the association matrix is updated by matrix multiplication for minimizing the effect of the false negative data. The heterogeneous lncRNA-disease network is finally analyzed using an unbalanced random walk method for predicting the potential lncRNA-disease associations. The LUNCRW model is therefore developed for predicting potential lncRNA-disease associations. The area under the curve (AUC) values of the LUNCRW model in leave-one-out cross-validation and five-fold cross-validation were 0.951 and 0.9486 ± 0.0011, respectively. Data from published case studies on three diseases, including squamous cell carcinoma, hepatocellular carcinoma, and renal cell carcinoma, confirmed the predictive potential of the LUNCRW model. Altogether, the findings indicated that the performance of the LUNCRW method is superior to that of existing methods in predicting potential lncRNA-disease associations.

Vocal cord lesions in representative autoimmune diseases.

OBJECTIVE: The aim of this study was to describe the clinical features of vocal cord lesions in patients with representative autoimmune diseases including systemic lupus erythematosus (SLE) and rheumatoid arthritis (RA). METHOD: A total of 31 SLE/RA patients (14 SLE and 17 RA) complicated with vocal cord lesions (SLE/RA-VC group) who had been admitted to Peking Union Medical College Hospital were retrieved from the electronic registration system. Ninety-three age and sex-matched SLE/RA patients (42 SLE and 51 RA) without vocal cord lesions (SLE/RA-nVC group) admitted during the same period were chosen randomly as controls. Medical files were reviewed and clinical data collected for comparisons. RESULTS: Vocal cord paralysis (n = 12, 38.7%) and vocal cord mass (n = 14, 45.2%) were the most common types of vocal cord lesions in this cohort. Unilateral lesions were more common than bilateral lesions (67.8% vs 32.3%) and the two sides were affected equally. Two cases of vocal cord bamboo node lesion were observed in SLE-VC group even as an initial manifestation and SLE-VC group had a slightly higher disease activity index (SLEDAI-2K) than their control counterparts (18.56 ± 8.23 vs 13.63 ± 5.89, p = 0.041). The RA-VC group had less pulmonary interstitial disease (29.4% vs 63%, p = 0.017) and lower CRP levels (p = 0.006) than their controls. As for the treatment, 71% of SLE/RA-VC patients had received glucocorticoids and immunosuppressants and 30% had undergone surgery. 45.2% of SLE/RA-VC patients got improvement at the time of discharge. CONCLUSIONS: The association of vocal cord lesions with disease activity can be observed in SLE patients but not in RA patients. Vocal cord lesions in SLE/RA patients should be considered as a part of the systemic involvement and should be treated accordingly. Key Points • Vocal cord paralysis and vocal cord mass were the main types of vocal cord lesions in patients with SLE/RA. • Vocal cord lesions in SLE patients may associate with disease activity and vocal cord bamboo node lesions could be an initial manifestation. • Glucocorticoid and immunosuppressants could be effective for vocal cord lesions in SLE/RA patients.

High expression of YKT6 associated with progression and poor prognosis of hepatocellular carcinoma.

BACKGROUND AND OBJECTIVE: The N-ethylmaleimide-sensitive fusion protein attachment protein receptor YKT6 is a key protein that controls the release of exosomes, was reported to play important roles in multiple cancers. However, the role of YKT6 in hepatocellular carcinoma (HCC) is still unknown. METHODS: Here we first used bioinformatics tools to analyze the YKT6 mRNA expression in HCC. In addition, we retrospectively collected 330 cancer tissue specimens from HCC patients and 180 para-cancerous tissue specimens, and detected YKT6 expression using immunohistochemical staining. Then the relationship between YKT6 expression and the clinical characteristics of HCC was analyzed, Kaplan-Meier analysis and Cox regression model were also performed to evaluate the impact of YKT6 expression on prognosis of HCC. Protein-protein interaction network of YKT6, and the gene enrichment analysis (GSEA) database were used to predict possible signal pathways regulated by YKT6 in HCC. RESULTS: The high expression rate of YKT6 in HCC (72.40%, 239/330) was higher than that in adjacent tissues (17.80%, 32/180, p < .001), and high expression of YKT6 was correlated with tumor size (p = 0.002), Edmondson Grade (p < .001), metastasis (p < .001), microvascular invasion (p = .005), AFP level (p = .002). Kaplan-Meier survival analysis showed that HCC patients with high YKT6 expression level had poorer prognosis. Meanwhile, multivariate Cox regression analysis showed that Edmondson grade (p = .009), metastasis (p = .049), YKT6 expression (p = .037) are independent risk factors for poor prognosis of HCC. Conclusions: Our results suggested that the upregulated expression of YKT6 is closely related to the progression HCC, which may be used as a potential biomarker for poor prognosis in HCC.

BRD4 and PIN1 gene polymorphisms are associated with high pulse pressure risk in a southeastern Chinese population.

BACKGROUND: BRD4 and PIN1 have been described to be involved in inflammation and vascular endothelial cell dysfunction, which in turn may increase pulse pressure. HYPOTHESIS: Genetic mutations within the BRD4 and PIN1 genes could affect the risk of high pulse pressure. METHODS: A total of four single nucleotide polymorphisms (SNPs) (BRD4: rs4808278; PIN1: rs2233678, rs2287838, and rs2233682) were genotyped in a cohort of 666 hypertensive patients and 232 normotensive controls with Chinese Han origin. Generalized multifactor dimensionality reduction (GMDR) was used to screen the best interaction combination among the four SNPs within the BRD4 and PIN1 genes and diabetes. Logistic regression analysis was performed to calculate the odds ratio (ORs) (95% confidence interval (CI)) for the association between the four SNPs. RESULTS: Adjusted for age, weight, waist circumference, drinking, smoking, hypertension, and diabetes, high pulse pressure risk was significantly higher for carriers with the rs4808278-TT genotype in BRD4 than those with wild genotypes (OR: 0.400, 95% CI: 0.217-0.737, P* < 0.05). However, we did not find any significant association of rs2233678, rs2287838, and rs2233682 in PIN1 with high pulse pressure susceptibility after covariate adjustment. GMDR analysis indicated a significant three-locus model (P = 0.0107) involving rs4808278, rs2233678, and diabetes, the cross-validation consistency of the three-locus models was 9/10, and the testing accuracy was 57.47%. CONCLUSIONS: Genetic mutations within BRD4 (rs4808278) could affect the susceptibility to high pulse pressure in a southeastern Chinese population.