Identification of a Potential MiRNA-mRNA Regulatory Network for Osteoporosis by Using Bioinformatics Methods: A Retrospective Study Based on the Gene Expression Omnibus Database

A Machine Learning Framework for Screening Plasma Cell-Associated Feature Genes to Estimate Osteoporosis Risk and Treatment Vulnerability

Osteoporosis, in which bones become fragile owing to low bone density and impaired bone mass, is a global public health concern. Bone mineral density (BMD) has been extensively evaluated for the diagnosis of low bone mass and osteoporosis. Circulating monocytes play an indispensable role in bone destruction and remodeling. This work proposed a machine learning-based framework to investigate the impact of circulating monocyte-associated genes on bone loss in osteoporosis patients. Females with discordant BMD levels were included in the GSE56815, GSE7158, GSE7429, and GSE62402 datasets. Circulating monocyte types were quantified via CIBERSORT, with subsequent selection of plasma cell-associated DEGs. Generalized linear models, random forests, extreme gradient boosting (XGB), and support vector machines were adopted for feature selection. Artificial neural networks and nomograms were subsequently constructed for osteoporosis diagnosis, and the molecular machinery underlying the identified genes was explored. SVM outperformed the other tuned models; thus, the expression of several genes (DEFA4, HLA-DPB1, LCN2, HP, and GAS7) associated with osteoporosis were determined. ANNs and nomograms were proposed to robustly distinguish low and high BMDs and estimate the risk of osteoporosis. Clozapine, aspirin, pyridoxine, etc. were identified as possible treatment agents. The expression of these genes is extensively posttranscriptionally regulated by miRNAs and m6A modifications. Additionally, they participate in modulating key signaling pathways, e.g., autophagy. The machine learning framework based on plasma cell-associated feature genes has the potential for estimating personalized risk stratification and treatment vulnerability in osteoporosis patients.

Prognostic analysis and validation of diagnostic marker genes in patients with osteoporosis

Backgrounds

As a systemic skeletal dysfunction, osteoporosis (OP) is characterized by low bone mass and bone microarchitectural damage. The global incidences of OP are high.

Methods

Data were retrieved from databases like Gene Expression Omnibus (GEO), GeneCards, Search Tool for the Retrieval of Interacting Genes/Proteins (STRING), Gene Expression Profiling Interactive Analysis (GEPIA2), and other databases. R software (version 4.1.1) was used to identify differentially expressed genes (DEGs) and perform functional analysis. The Least Absolute Shrinkage and Selection Operator (LASSO) logistic regression and random forest algorithm were combined and used for screening diagnostic markers for OP. The diagnostic value was assessed by the receiver operating characteristic (ROC) curve. Molecular signature subtypes were identified using a consensus clustering approach, and prognostic analysis was performed. The level of immune cell infiltration was assessed by the Cell-type Identification by Estimating Relative Subsets of RNA Transcripts (CIBERSORT) algorithm. The hub gene was identified using the CytoHubba algorithm. Real-time fluorescence quantitative PCR (RT-qPCR) was performed on the plasma of osteoporosis patients and control samples. The interaction network was constructed between the hub genes and miRNAs, transcription factors, RNA binding proteins, and drugs.

Results

A total of 40 DEGs, eight OP-related differential genes, six OP diagnostic marker genes, four OP key diagnostic marker genes, and ten hub genes (TNF, RARRES2, FLNA, STXBP2, EGR2, MAP4K2, NFKBIA, JUNB, SPI1, CTSD) were identified. RT-qPCR results revealed a total of eight genes had significant differential expression between osteoporosis patients and control samples. Enrichment analysis showed these genes were mainly related to MAPK signaling pathways, TNF signaling pathway, apoptosis, and Salmonella infection. RT-qPCR also revealed that the MAPK signaling pathway (p38, TRAF6) and NF-kappa B signaling pathway (c-FLIP, MIP1β) were significantly different between osteoporosis patients and control samples. The analysis of immune cell infiltration revealed that monocytes, activated CD4 memory T cells, and memory and naïve B cells may be related to the occurrence and development of OP.

Conclusions

We identified six novel OP diagnostic marker genes and ten OP-hub genes. These genes can be used to improve the prognostic of OP and to identify potential relationships between the immune microenvironment and OP. Our research will provide insights into the potential therapeutic targets and pathogenesis of osteoporosis.