Computational and Mathematical Methods in Medicine

Spinal cord injury (SCI) is a serious disorder of the central nervous system with a high disability rate. Long noncoding RNAs (lncRNAs) are reported to mediate many biological processes. The aim of this study was to explore lncRNA and mRNA expression profiles and functional networks after SCI. Differentially expressed genes between SCI model rats and sham controls were identified by microarray assays and analyzed by functional enrichment. Key lncRNAs were identified using a support vector machine- (SVM-) recursive feature elimination (RFE) algorithm. A trans and cis regulation model was used to analyze the regulatory relationships between lncRNAs and their targets. An lncRNA-related ceRNA network was established. We identified 5465 differentially expressed lncRNAs (DE lncRNAs) and 8366 differentially expressed mRNAs (DE mRNAs) in the SCI group compared with the sham group (, ). Four genes were confirmed by qRT-PCR which were consistent with the microarray data. GSEA analysis showed that most marked changes occurred in pathways related to immune inflammation and nerve cell function, including cytokine-cytokine receptor interaction, neuroactive ligand-receptor interaction, and GABAergic synapse. Enrichment analysis identified 30 signaling pathways, including those associated with immune inflammation response. A total of 40 key lncRNAs were identified using the SVM-RFE algorithm. A key lncRNA-mRNAs coexpression network was generated for 230 951 lncRNA-mRNA pairs with half showing positive correlations. Several key DE lncRNAs were predicted to have “cis”- or “trans”-regulated target genes. The transcription factors, Sp1, JUN, and SOX10, may regulate the interaction between XR_001837123.1 and ETS 1. In addition, five pairs of ceRNA regulatory sequences were constructed. Many mRNAs and lncRNAs were found to be dysregulated after SCI. Bioinformatic analysis showed that DE lncRNAs may play crucial roles in SCI. It is anticipated that these findings will provide new insights into the underlying mechanisms and potential therapeutic targets for SCI.

Overall survival (OS) in cancer is crucial for cancer treatment. Many machine learning methods have been applied to predict OS, but there are still the challenges of dealing with multiview data and overfitting. To overcome these problems, we propose a multiview deep forest (MVDF) in this paper. MVDF can learn the features of each view and fuse them with integrated learning and multiple kernel learning. Then, a gradient boost forest based on the information bottleneck theory is proposed to reduce redundant information and avoid overfitting. In addition, a pruning strategy for a cascaded forest is used to limit the impact of outlier data. Comprehensive experiments have been carried out on a data set from West China Hospital of Sichuan University and two public data sets. Results have demonstrated that our method outperforms the compared methods in predicting overall survival.

Background. NADH dehydrogenase (ubiquinone) 1 alpha subcomplex assembly factor 2 (NDUFAF2) acts as a molecular chaperone for the assembly of complex I on the mitochondrial membrane, which is involved in the transfer of electrons in the respiratory chain. However, whether NDUFAF2 plays a role in lung adenocarcinoma (LUAD) is largely unexplored. Methods. Expression profiles were obtained from the TCGA and GEO databases and integrated via R3.6.3 and several bioinformatics platforms. Western blotting analysis and immunohistochemistry staining were used to examine the expressions of NDUFAF2 in clinical samples. Moreover, the diagnostic and prognostic value of NDUFAF2 expression level was also assessed. GO, KEGG, and gene set enrichment analysis (GSEA) were adopted to investigate NDUFAF2-related molecular functions, signaling pathways, and life activity processes. Results. NDUFAF2 was predominantly expressed in LUAD, and it is identified as a promising biomarker in the diagnosis of LUAD and its prognostic prediction. Overexpression of NDUFAF2 was correlated with N stage, T stage, and pathologic stage in LUAD, leading to worse overall survival (OS). Besides, the level of NDUFAF2 was independently associated with OS through a multivariate Cox analysis (, 95% (1.086-2.177), ). GO analysis revealed enrichment in innate immune response in mucosa and mucosal immune response, and GSEA indicated enrichment in G2_M_checkpoints, DNA replication, diseases of mitotic cell cycle, retinoblastoma gene in cancer, cell cycle pathway, and cell cycle. Furthermore, the expression level of NDUFAF2 was negatively correlated with infiltration levels of Tem, Tcm, NK CD56^bright cells, and B cells. In contrast, the expression level of NDUFAF2 was positively correlated with the infiltration level of DCs and Th2 cells in LUAD patients. Conclusions. Collectively, NDUFAF2 is a promising independent prognostic biomarker and target in LUAD. In addition, NDUFAF2 might affect the prognosis of LUAD via DNA replication, diseases of mitotic cell cycle, cell cycle pathway, and cell cycle.

Objective. The objective of this study was to explore the medicinal properties of herbal medicines that can interfere with the copper death pathway. Methods. The Human Gene Database, Chemical Interactions in Comparative Toxicogenomics Database, Encyclopedia of Traditional Chinese Medicine, China Medical Information Platform, and Cytoscape software were used to find target and chemicals that interfere with copper death targets, as well as herbal medicines containing these chemicals and their four natures and five flavors (basic properties of herbal medicines). Results. 27 copper death-related targets were finally retrieved, as well as 2143 chemicals that could interfere with them, including 180 herbal compounds. The compounds with the highest degree values (number of nodes connected to this node) were folic acid, resveratrol, and quercetin. The 180 compounds were related to 278 herbs; those with the highest degree values (number of nodes connected to this node) were Jujubae Fructus, Ginkgo biloba L, and Acanthopanax senticosus. The 27 copper death targets were indirectly associated with 278 herbs; those with the highest degree values (number of nodes connected to this node) were Achyranthis Bidentatae Radix, Polygonum cuspidatum Sieb. et Zucc, and Mori Folium. Among the 278 herbs, 6 had incomplete information. A pharmacological analysis showed that among the 272 Chinese herbs, the most frequent meridians were the liver (133), lung (104), and spleen (91). Of the four natures, the most frequent were cold (73), warm (68), and flat (45). Of the five flavors, the most frequent were bitter (165), pungent (116), and sweet (99). Conclusion. This study preliminarily discussed the material basis and medicinal properties of herbs that can intervene in copper death, which can provide reference for the theoretical discussion, drug development, and clinical research of Chinese medicine regulating copper death.

In recent years, with the increasing incidence of cancer, regular physical examination is an important way to find cancer. Nuclear screening is an important method for the diagnosis of gastrointestinal diseases, but it is challenging in the face of small and fuzzy gastrointestinal images. Different from traditional medical objects, pathological slice images are mostly blurry and tiny, which is somewhat difficult to detect and segment. The traditional diagnostic method lacks rapid quantitative analysis and has a certain delay in medical diagnosis, and traditional image processing uses morphological features and pixel distribution to extract features; it is often difficult to achieve the desired effect on small blurry images. This paper proposes a small, microfuzzy pathology detection algorithm based on the attention mechanism; the YOLOv5 is improved under small and micro fuzzy scenarios of the detection of cancer cells in the full field of digital pathology and tests it in the gastric cancer slice dataset. The network structure is improved, and the ability to learn features on small and micro targets is enhanced according to the law of feature distribution. Spatial and channel changes in network attention and attention weight distribution. In the deep blur scenario, the attention mechanism is added to optimize its recognition ability, and the test result shows F1_score is 0.616, and the mAP is 0.611, which can provide the decision support for clinical judgment.

Background. Ankylosing spondylitis (AS) and inflammatory bowel disease (IBD) are both autoimmune diseases, and they often occur together in clinical practice, but the pathogenesis is unclear. This study is aimed at identifying the hub genes and explore the related immune molecular mechanisms between AS and IBD by bioinformatics analysis. Methods. From the public Gene Expression Omnibus (GEO) database, the AS and IBD datasets (GSE73754, GSE59071, GSE25101, and GSE36807) were obtained. The immune cell infiltration in the peripheral blood tissues of GSE73754 and GSE59071 was assessed using the CIBERSORT algorithm. Then, we used the Weighted Gene Coexpression Network Analysis (WGCNA) to identify the Differentially Expressed Genes (DEGs) related to AS and IBD. Then, the immune genes from the ImmPort database intersected with the DEGs to obtain hub genes. The Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyzed the functional correlation of hub genes. Then, hub genes were verified in GSE25101 and GSE36807. The clusterProfiler software and Gene Set Enrichment Analysis (GSEA) were used to conduct functional enrichment and pathway enrichment studies. Finally, the diagnostic efficacy was assessed using Receiver Operating Characteristic (ROC) curve analysis. Results. The analysis of immune characteristics showed that both AS and IBD were related to immunity, and neutrophils were positively correlated in both diseases. Nine coexpressed genes, including FCGRT, S100A11, IFNGR1, NFKBIZ, JAK2, LYN, PLAUR, ADM, and IL1RN, were linked to immune cells. The GO and KEGG analyses results showed that enrichment analysis was mainly related to cell transport and migration. Finally, the ROC curve was verified with the validation set, and it was found that PLAUR has clinical diagnostic significance and the most excellent specificity and sensitivity, respectively. Conclusions. PLAUR (uPAR) is a promising biomarker and will be an underlying genetic biomarker for diagnosing AS comorbid IBD. Inflammation and immunological modulation mediated by neutrophil infiltration were important in the development of AS and IBD and may be diagnostic and therapeutic targets.