Spina bifida is a common neural tube defect (NTD) accounting for 5-10% of perinatal mortalities. As a polygenic disease, spina bifida is caused by a combination of genetic and environmental factors, for which the precise molecular pathogenesis is still not systemically understood. In the present study, we aimed to identify the related gene module that might play a vital role in the occurrence and development of spina bifida by using weighted gene co-expression network analysis (WGCNA). Transcription profiling according to an array of human amniocytes from patients with spina bifida and healthy controls was downloaded from the Gene Expression Omnibus database. First, outliers were identified and removed by principal component analysis (PCA) and sample clustering. Then, genes in the top 25% of variance in the GSE4182 dataset were then determined in order to explore candidate genes in potential hub modules using WGCNA. After data preprocessing, 5407 genes were obtained for further WGCNA. Highly correlated genes were divided into nineteen modules. Combined with a co-expression network and significant differentially expressed genes, 967 candidate genes were identified that may be involved in the pathological processes of spina bifida. Combined with our previous microRNA (miRNA) microarray results, we constructed an miRNA-mRNA network including four miRNAs and 39 mRNA among which three key genes were, respectively, linked to two miRNA-associated gene networks. Following the verification of qRT-PCR and KCND3 was upregulated in the spina bifida. KCND3 and its related miR-765 and miR-142-3p are worthy of further study. These findings may be conducive for early detection and intervention in spina bifida, as well as be of great significance to pregnant women and clinical staff.
Keywords: bioinformatics analysis; hub genes; pathological process; spina bifida; weighted gene co-expression network analysis.
Copyright © 2020 Li, Feng and Yuan.