Search

ObjectiveThis study aimed to construct a machine learning model using clinical variables and ultrasound radiomics features for the prediction of the benign or malignant nature of pancreatic tumors.Methods242 pancreatic tumor patients who were hospitalized at the First Affiliated Hospital of Guangxi Medical University between January 2020 and June 2023 were included in this retrospective study. The patients were randomly divided into a training cohort (n=169) and a test cohort (n=73). We collected 28 clinical features from the patients. Concurrently, 306 radiomics features were extracted from the ultrasound images of the patients’ tumors. Initially, a clinical model was constructed using the logistic regression algorithm. Subsequently, radiomics models were built using SVM, random forest, XGBoost, and KNN algorithms. Finally, we combined clinical features with a new feature RAD prob calculated by applying radiomics model to construct a fusion model, and developed a nomogram based on the fusion model.ResultsThe performance of the fusion model surpassed that of both the clinical and radiomics models. In the training cohort, the fusion model achieved an AUC of 0.978 (95% CI: 0.96–0.99) during 5-fold cross-validation and an AUC of 0.925 (95% CI: 0.86–0.98) in the test cohort. Calibration curve and decision curve analyses demonstrated that the nomogram constructed from the fusion model has high accuracy and clinical utility.ConclusionThe fusion model containing clinical and ultrasound radiomics features showed excellent performance in predicting the benign or malignant nature of pancreatic tumors.

IntroductionHepatocellular carcinoma (HCC) is an aggressive malignancy with steadily increasing incidence rates worldwide and poor therapeutic outcomes. Studies show that metabolic reprogramming plays a key role in tumor genesis and progression. In this study, we analyzed the metabolic heterogeneity of epithelial cells in the HCC and screened for potential biomarkers.MethodsThe hepatic single-cell RNA sequencing (scRNA-seq) datasets of HCC patients and healthy controls were obtained from the Gene Expression Omnibus (GEO) database. Based on data intergration and measurement of differences among groups, the metabolic epithelial cell subpopulations were identified. The single-cell metabolic pathway was analyzed and the myeloid subpopulations were identified. Cell-cell interaction analysis and single-cell proliferation analysis were performed. The gene expression profiles of HCC patients were obtained from the GSE14520 dataset of GEO and TCGA-LIHC cohort of the UCSC Xena website. Immune analysis was performed. The differentially expressed genes (DEGs) were identified and functionally annotated. Tumor tissues from HCC patients were probed with anti-ALDOA, anti-CD68, anti-CD163, anti-CD4 and anti-FOXP3 antibodies. Results We analyzed the scRNA-seq data from 48 HCC patients and 14 healthy controls. The epithelial cells were significantly enriched in HCC patients compared to the controls (p = 0.011). The epithelial cells from HCC patients were classified into two metabolism-related subpopulations (MRSs) – pertaining to amino acid metabolism (MRS1) and glycolysis (MRS2). Depending on the abundance of these metabolic subpopulations, the HCC patients were also classified into the MRS1 and MRS2 subtype distinct prognoses and immune infiltration. The MRS2 group had significantly worse clinical outcomes and more inflamed tumor microenvironment (TME), as well as a stronger crosstalk between MRS2 cells and immune subpopulations that resulted in an immunosuppressive TME. We also detected high expression levels of ALDOA in the MRS2 cells and HCC tissues. In the clinical cohort, HCC patients with higher ALDOA expression showed greater enrichment of immunosuppressive cells including M2 macrophages and T regulatory cells.DiscussionThe glycolytic subtype of HCC cells with high ALDOA expression is associated with an immunosuppressive TME and predicts worse clinical outcomes, providing new insights into the metabolism and prognosis of HCC.

CircRNA is a new type of non-coding RNA with a closed loop structure. More and more biological experiments show that circRNA plays important roles in many diseases by regulating the target genes of miRNA. Therefore, correct identification of the potential interaction between circRNA and miRNA not only helps to understand the mechanism of the disease, but also contributes to the diagnosis, treatment, and prognosis of the disease. In this study, we propose a model (IIMCCMA) by using network embedding and matrix completion to predict the potential interaction of circRNA-miRNA. Firstly, the corresponding adjacency matrix is constructed based on the experimentally verified circRNA-miRNA interaction, circRNA-cancer interaction, and miRNA-cancer interaction. Then, the Gaussian kernel function and the cosine function are used to calculate the circRNA Gaussian interaction profile kernel similarity, circRNA functional similarity, miRNA Gaussian interaction profile kernel similarity, and miRNA functional similarity. In order to reduce the influence of noise and redundant information in known interactions, this model uses network embedding to extract the potential feature vectors of circRNA and miRNA, respectively. Finally, an improved inductive matrix completion algorithm based on the feature vectors of circRNA and miRNA is used to identify potential interactions between circRNAs and miRNAs. The 10-fold cross-validation experiment is utilized to prove the predictive ability of the IIMCCMA. The experimental results show that the AUC value and AUPR value of the IIMCCMA model are higher than other state-of-the-art algorithms. In addition, case studies show that the IIMCCMA model can correctly identify the potential interactions between circRNAs and miRNAs.

Osteosarcoma is a kind of primary malignant tumor of bone. In recent years, its therapeutic effect and prognostic survival are dissatisfactory. The tumor immune microenvironment (TIME) reflects immune status of patients, but it is little known in osteosarcoma. Therefore, this study attempts to conduct a comprehensive analysis to explore TIME of osteosarcoma and identify TIME-related subtypes for clinical management and treatment. We successfully established two novel tumor immune infiltration clusters (TIIC) which are characterized by difference of microenvironment and immune-related biological processes. High tumor immune infiltration cluster (H-TIIC) subtypes with higher immune infiltration score shows a better overall survival. Further, the two immune subtypes are shown to differ in immunotherapy and chemotherapy. The results would be helpful for clinical decision in osteosarcoma.

Osteoarthritis (OA) is a common chronic degenerative arthritis. Its treatment options are very limited. At present, hypoxia is a prominent factor in OA. This study aimed to re-explore the mechanism between hypoxia and OA, which provides new insights into the diagnosis and therapy of OA. We acquired the OA-related expression profiles of GSE48556, GSE55235, and GSE55457 for our analysis. Using gene set variation analysis (GSVA), we found significant differences in hypoxia. These differences result from multiple pathways, such as the p53 signaling pathway, cell senescence, the NF-kappa B signaling pathway, Ubiquitin-mediated proteolysis, and apoptosis. Meanwhile, the single-sample gene set enrichment analysis (ssGSEA) showed that hypoxia was significantly associated with the level of immune cell infiltration in the immune microenvironment. Thus, we believe that hypoxia is useful for the diagnosis and treatment of OA. We successfully constructed a novel hypoxia-related index (HRI) based on seven hypoxia-related genes (ADM, CDKN3, ENO1, NDRG1, PGAM1, SLC2A1, VEGFA) by least absolute shrinkage and binary logistic regression of the generalized linear regression. HRI showed potential for improving OA diagnosis through receiver operation characteristic (ROC) analysis (AUC training cohort = 0.919, AUC testing cohort = 0.985). Moreover, we found that celastrol, droxinostat, torin-2, and narciclasine may be potential therapeutic compounds for OA based on the Connectivity Map (CMap). In conclusion, hypoxia is involved in the development and progression of OA. HRI can improve diagnosis and show great potential in clinical application. Celastrol, droxinostat, torin-2, and narciclasine may be potential compounds for the treatment of OA patients.