Integrative Analysis of Genomics and Transcriptome Data to Identify Regulation Networks in Female Osteoporosis

Abstract

Background: Osteoporosis is a highly heritable skeletal muscle disease. However, the genetic mechanisms mediating the pathogenesis of osteoporosis remain unclear. Accordingly, in this study, we aimed to clarify the transcriptional regulation and heritability underlying the onset of osteoporosis.Methods: Transcriptome gene expression data were obtained from the Gene Expression Omnibus database. Microarray data from peripheral blood monocytes of 73 Caucasian women with high and low bone mineral density (BMD) were analyzed. Differentially expressed messenger RNAs (mRNAs) and long non-coding RNAs (lncRNAs) were identified. Differences in BMD were then attributed to several gene modules using weighted gene co-expression network analysis (WGCNA). LncRNA/mRNA regulatory networks were constructed based on the WGCNA and subjected to functional enrichment analysis.Results: In total, 3,355 mRNAs and 999 lncRNAs were identified as differentially expressed genes between patients with high and low BMD. The WGCNA yielded three gene modules, including 26 lncRNAs and 55 mRNAs as hub genes in the blue module, 36 lncRNAs and 31 mRNAs as hub genes in the turquoise module, and 56 mRNAs and 30 lncRNAs as hub genes in the brown module. JUN and ACSL5 were subsequently identified in the modular gene network. After functional pathway enrichment, 40 lncRNAs and 16 mRNAs were found to be related to differences in BMD. All three modules were enriched in metabolic pathways. Finally, mRNA/lncRNA/pathway networks were constructed using the identified regulatory networks of lncRNAs/mRNAs and pathway enrichment relationships.Conclusion: The mRNAs and lncRNAs identified in this WGCNA could be novel clinical targets in the diagnosis and management of osteoporosis. Our findings may help elucidate the complex interactions between transcripts and non-coding RNAs and provide novel perspectives on the regulatory mechanisms of osteoporosis.

Association between Sensitivity to Thyroid Hormone Indices and Bone Mineral Density in US Males

“…Osteoporosis is a metabolic bone disease characterized by the degeneration of bone tissue microstructure and decreased bone mineral density. The early clinical manifestations of osteoporosis are subtle, but many patients often have clinical manifestations such as spinal deformity, bone pain, and fractures in the later stage of osteoporosis [ 1 , 2 ]. With the aging population, osteoporosis has already become an increasingly significant public health problem [ 3 ].…”

Screening of Important Markers in Peripheral Blood Mononuclear Cells to Predict Female Osteoporosis Risk Using LASSO Regression Algorithm and SVM Method

“…In the past few decades, a massive number of researches have been made in exploring key genes in osteoporosis, especially for genes which involved in bone metabolism. [6][7][8][9] Many potentially important genes were identified by comparing high hip and low hip BMD samples in different cohorts but there still lack comprehensive analysis and detailed exploration about marker genes for this disease.…”

“…3 Peripheral blood mononuclear (PBMs) cells may act as precursors for osteoclasts (bone resorbing cells) 4 especially for the adult peripheral skeleton (eg, femur as one of the most important skeletal sites) where circulating mononuclear cells provide the only source of osteoclast precursors, making PBMs a suitable cellular model for the study of osteoporosis. 5 There have been a massive number of studies analyzing genes for dysregulation between high and low BMD, 6 - 9 but there is still a lack of validated markers, while most of the studies were limited to a single dataset without further mechanistic exploration of the screened genes. Therefore, in this analysis, we combined multiple datasets, used LASSO regression for screening of marker features between high and low BMD, and used the validation set data to construct SVM model to evaluate the diagnostic accuracy of the feature genes, and further combined with the upstream miRNAs of the feature genes and transcription factor regulatory mechanisms to explore downstream functions.…”

Identification and Functional Analysis of Long Non-coding RNAs in Human Pulmonary Microvascular Endothelial Cells Subjected to Cyclic Stretch

“…The microarray data were pre-processed using the Robust Multi-array Average (RMA) method for background-corrected, normalization, and summary on the GCBI bioinformatics platform (Kong et al, 2016 ; Huang et al, 2018 ). Subsequently, Significant Analysis of Microarray (SAM) algorithm was used to identify differentially expressed lncRNAs (DELs) and differentially expressed genes (DEGs) with the fold change cutoff of 1.2 and p -value cutoff of 0.05 according to the prior reported literatures (Jia and Zhai, 2019 ; Li et al, 2020 ; Zhang et al, 2020 ). Hierarchical clustering was performed to distinguish the different gene clustering patterns on the same platform.…”