Search

ObjectiveIschemic stroke (IS) with subsequent cerebrocardiac syndrome (CCS) has a poor prognosis. We aimed to investigate electrocardiogram (ECG) changes after IS with artificial intelligence (AI).MethodsWe collected ECGs from a healthy population and patients with IS, and then analyzed participant demographics and ECG parameters to identify abnormal features in post-IS ECGs. Next, we trained the convolutional neural network (CNN), random forest (RF) and support vector machine (SVM) models to automatically detect the changes in the ECGs; Additionally, We compared the CNN scores of good prognosis (mRS ≤ 2) and poor prognosis (mRS > 2) to assess the prognostic value of CNN model. Finally, we used gradient class activation map (Grad-CAM) to localize the key abnormalities.ResultsAmong the 3506 ECGs of the IS patients, 2764 ECGs (78.84%) led to an abnormal diagnosis. Then we divided ECGs in the primary cohort into three groups, normal ECGs (N-Ns), abnormal ECGs after the first ischemic stroke (A-ISs), and normal ECGs after the first ischemic stroke (N-ISs). Basic demographic and ECG parameter analyses showed that heart rate, QT interval, and P-R interval were significantly different between 673 N-ISs and 3546 N-Ns (p < 0.05). The CNN has the best performance among the three models in distinguishing A-ISs and N-Ns (AUC: 0.88, 95%CI = 0.86-0.90). The prediction scores of the A-ISs and N-ISs obtained from the all three models are statistically different from the N-Ns (p < 0.001). Futhermore, the CNN scores of the two groups (mRS > 2 and mRS ≤ 2) were significantly different (p < 0.05). Finally, Grad-CAM revealed that the V4 lead may harbor the highest probability of abnormality.ConclusionOur study showed that a high proportion of post-IS ECGs harbored abnormal changes. Our CNN model can systematically assess anomalies in and prognosticate post-IS ECGs.

As cancer mortality is high in most regions of the world, early screening of cancer has become increasingly important. Minimally invasive screening programs that use peripheral blood mononuclear cells (PBMCs) are a new and reliable strategy that can achieve early detection of tumors by identifying marker genes. From 797 datasets, four (GSE12771, GSE24536, GSE27562, and GSE42834) including 428 samples, 236 solid tumor cases, and 192 healthy controls were chosen according to the inclusion criteria. A total of 285 genes from among 440 reported genes were selected by meta-analysis. Among them, 4 of the top significantly differentially expressed genes (ANXA1, IFI44, IFI44L, and OAS1) were identified as marker genes of PBMCs. Pathway enrichment analysis identified, two significant pathways, the 'primary immunodeficiency' pathway and the 'cytokine-cytokine receptor interaction' pathway. Protein- protein interaction (PPI) network analysis revealed the top 27 hubs with a degree centrality greater than 23 to be hub genes. We also identified 3 modules in Molecular Complex Detection (MCODE) analysis: Cluster 1 (related to ANXA1), Cluster 2 (related to IFI44 and IFI44L) and Cluster 3 (related to OAS1). Among the 4 marker genes, IFI44, IFI44L, and OAS1 are potential diagnostic biomarkers, even though their results were not as remarkable as those for ANXA1 in our study. ANXA1 is involved in the immunosuppressive mechanism in tumor-bearing hosts and may be used in a new strategy involving the use of the host's own immunity to achieve tumor suppression.

BACKGROUND:Hepatocellular carcinoma (HCC) is one of the most lethal and malignant tumours worldwide. New therapeutic targets for HCC are urgently needed. CYCLOPS (copy number alterations yielding cancer liabilities owing to partial loss) genes have been noted to be associated with cancer-targeted therapies. Therefore, we intended to explore the effects of the CYCLOPS gene RBM17 on HCC oncogenesis to determine if it could be further used for targeted therapy. METHODS:We collected data on 12 types of cancer from the Cancer Genome Atlas (TCGA) and Gene Expression Omnibus (GEO) queries for comparison with adjacent non-tumour tissues. RBM17 expression levels, clinicopathological factors and survival times were analysed. RNAseq data were downloaded from the Encyclopaedia of DNA Elements database for molecular mechanism exploration. Two representative HCC cell models were built to observe the proliferation capacity of HCC cells when RBM17 expression was inhibited by shRBM17. Cell cycle progression and apoptosis were also examined to investigate the pathogenesis of RBM17. RESULTS:Based on 6,136 clinical samples, RBM17 was markedly overexpressed in most cancers, especially HCC. Moreover, data from 442 patients revealed that high RBM17 expression levels were related to a worse prognosis. Overexpression of RBM17 was related to the iCluster1 molecular subgroup, TNM stage, and histologic grade. Pathway analysis of RNAseq data suggested that RBM17 was involved in mitosis. Further investigation revealed that the proliferation rates of HepG2 (P = 0.003) and SMMC-7721 (P = 0.030) cells were significantly reduced when RBM17 was knocked down. In addition, RBM17 knockdown also arrested the progression of the cell cycle, causing cells to halt at the G2/M phase. Increased apoptosis rates were also found in vitro. CONCLUSION:These results suggest that RBM17 is a potential therapeutic target for HCC treatment.

SWItch/Sucrose NonFermentable (SWI/SNF) is a set of multi-subunits chromatin remodeling complexes, playing important roles in a variety of biological processes. Loss-of-function mutations in the genes encoding SWI/SNF subunits have been reported in more than 20% of human cancers. Thus, it was widely considered as a tumor suppressor in the past decade. However, recent studies reported that some genes encoding subunits of SWI/SNF complexes were amplified and play oncogenic roles in human cancers. In present study, we summarized the genetic alteration spectrum of SWI/SNF complexes, and firstly systematically estimated both the copy number variations and point mutations of all 30 genes encoding the subunits in this complex. Additionally, the bioinformatics analyses were performed for two significantly amplified genes, ACTL6A and BRD9, to investigate their oncogenic roles in human cancers. Our findings may lay a foundation for the discovery of potential treatment targets in SWI/SNF complexes of cancers.

Recent evidence suggests that aberrant activation of Hedgehog (Hh) signaling by Gli transcription factors is characteristic of a variety of aggressive human carcinomas including ovarian cancer. Therefore, chemotherapeutic agents that inhibit activation of Gli transcription factors have emerged as promising novel therapeutic drugs for ovarian cancer.In this study, we show that activation of Hh signaling promoted cellular migration and invasion, whereas blockade of Hh signaling with GANT61 suppressed cellular migration and invasion in ovarian cancer cells. After treatment with GANT61, cDNA microarray analyses revealed changes in many genes such as Integrin β4 subunit (ITGB4), focal adhesion kinase (FAK), etc. Furthermore, ITGB4 expression was up-regulated by Sonic Hedgehog (Shh) ligand and down-regulated by Hh signaling inhibitor. The Shh-mediated ovarian cell migration and invasion was blocked by neutralizing antibodies to ITGB4. In addition, phosphorylations of FAK were increased by Shh and decreased by Hh signaling inhibitor. Inhibition of Gli1 expression using siRNA mimicked the effects of GANT61 treatment, supporting the specificity of GANT61. Further investigations showed that activation of FAK was required for Shh-mediated cell migration and invasion. Finally, we found that down-regulation of Gli reduced the expression of ITGB4 and the phosphorylated FAK, resulting in the inhibition of tumor growth in vivo.The Hh signaling pathway induces cell migration and invasion through ITGB4-mediated activation of FAK in ovarian cancer. Our findings suggest that the diminishment of crosstalk between phosphorylated FAK and ITGB4 due to the down-regulation of Gli family transcription factors might play a pivotal role for inhibiting ovarian cancer progression.