Construction of a combined random forest and artificial neural network diagnosis model to screening potential biomarker for hepatoblastoma

Liu, Shaowen; Zheng, Qipeng; Zhang, Ruifeng; Li, Tengfei; Zhan, Jianghua

doi:10.1007/s00383-022-05255-3

Cited by 2 publications

(1 citation statement)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Studies have reported that the expression of memory B-cell is significantly different in patients with ulcerative colitis compared to normal subjects ( Pararasa et al, 2019 ). In hepatoblastoma, the relative abundance of memory B-cell is positively correlated with disease severity ( Liu et al, 2022 ). In HSCR, abnormalities in gut microbial regulation usually cause HAEC, resulting in delayed thymic development.…”

Section: Discussionmentioning

confidence: 99%

Weighted gene co-expression network analysis and CIBERSORT screening of key genes related to m6A methylation in Hirschsprung’s disease

et al. 2023

View full text Add to dashboard Cite

Hirschsprung’s disease (HSCR) is a neural crest disease that results from the failure of enteric neural crest cells (ENCCs) to migrate to the corresponding intestinal segment. The RET gene, which regulates enteric neural crest cell proliferation and migration, is considered one of the main risk factors for HSCR and is commonly used to construct HSCR mouse models. The epigenetic mechanism of m6A modification is involved in HSCR. In this study, we analyzed the GEO database (GSE103070) for differentially expressed genes (DEGs) and focused on m6A–related genes. Comparing the RNA-seq data of Wide Type and RET Null, a total of 326 DEGs were identified, of which 245 genes were associated with m6A. According to the CIBERSORT analysis, the proportion of Memory B-cell in RET Null was significantly higher than that of Wide Type. Venn diagram analysis was used to identify key genes in the selected memory B-cell modules and DEGs associated with m6A. Enrichment analysis showed that seven genes were mainly involved in focal adhesion, HIV infection, actin cytoskeleton organization and regulation of binding. These findings could provide a theoretical basis for molecular mechanism studies of HSCR.

show abstract

Section: Discussionmentioning

confidence: 99%

Weighted gene co-expression network analysis and CIBERSORT screening of key genes related to m6A methylation in Hirschsprung’s disease

et al. 2023

View full text Add to dashboard Cite

show abstract

A case–control comparison of acute-phase peripheral blood gene expression in participants diagnosed with minor ischaemic stroke or stroke mimics

Moxon,

Calcino,

Kraeuter

et al. 2023

Hum Genomics

View full text Add to dashboard Cite

Background Past studies suggest that there are changes in peripheral blood cell gene expression in response to ischaemic stroke; however, the specific changes which occur during the acute phase are poorly characterised. The current study aimed to identify peripheral blood cell genes specifically associated with the early response to ischaemic stroke using whole blood samples collected from participants diagnosed with ischaemic stroke (n = 29) or stroke mimics (n = 27) following emergency presentation to hospital. Long non-coding RNA (lncRNA), mRNA and micro-RNA (miRNA) abundance was measured by RNA-seq, and the consensusDE package was used to identify genes which were differentially expressed between groups. A sensitivity analysis excluding two participants with metastatic disease was also conducted. Results The mean time from symptom onset to blood collection was 2.6 h. Most strokes were mild (median NIH stroke scale score 2.0). Ten mRNAs (all down-regulated in samples provided by patients experiencing ischaemic stroke) and 30 miRNAs (14 over-expressed and 16 under-expressed in participants with ischaemic stroke) were significantly different between groups in the whole cohort and sensitivity analyses. No significant over-representation of gene ontology categories by the differentially expressed genes was observed. Random forest analysis suggested a panel of differentially expressed genes (ADGRG7 and miRNAs 96, 532, 6766, 6798 and 6804) as potential ischaemic stroke biomarkers, although modelling analyses demonstrated that these genes had poor diagnostic performance. Conclusions This study provides evidence suggesting that the early response to minor ischaemic stroke is predominantly reflected by changes in the expression of miRNAs in peripheral blood cells. Further work in independent cohorts particularly in patients with more severe stroke is needed to validate these findings and investigate their clinical relevance.

show abstract

Construction of a combined random forest and artificial neural network diagnosis model to screening potential biomarker for hepatoblastoma

Cited by 2 publications

References 41 publications

Weighted gene co-expression network analysis and CIBERSORT screening of key genes related to m6A methylation in Hirschsprung’s disease

Weighted gene co-expression network analysis and CIBERSORT screening of key genes related to m6A methylation in Hirschsprung’s disease

A case–control comparison of acute-phase peripheral blood gene expression in participants diagnosed with minor ischaemic stroke or stroke mimics

Contact Info

Product

Resources

About