Related Experiment Video
Updated: Aug 19, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
The risk of COVID-19 can be predicted by a nomogram based on m6A-related genes
Lingling Lu1, Yijing Li2, Xiulan Ao3
1Fuzong Clinical Medical College of Fujian Medical University, The 900th hospital. No.156 Xierhuan Road, Fuzhou, Fujian 350025, China; Department of Hepatobiliary Disease, 900th Hospital of Joint Logistics Support Force, No.156 Xierhuan Road, Fuzhou, Fujian 350025, China.
Background:
The expression of m6A-related genes and their significance in COVID-19 patients are still unknown.
Methods:
The GSE177477 and GSE157103 datasets of the Gene Expression Omnibus were used to extract RNA-seq data. The expression of 26 m6A-related genes and immune cell infiltration in COVID-19 patients were analyzed. Finally, we built and validated a nomogram model to predict the risk of COVID-19 infection.
Results:
There were significant differences in 11 m6A regulatory factors between patients with COVID-19 and healthy individuals. The classification of disease subtypes based on m6A-related gene levels can be distinguished. COVID-19 patients in GSE177477 were classified into two categories based on m6A-related genes. The patients in cluster A were all symptomatic, while those in cluster B were asymptomatic. A significant correlation was also found between immune cells and m6A-related genes. Finally, seven m6A-related disease-characteristic genes, HNRNPA2B1, ELAVL1, RBM15, RBM15B, YTHDC1, HNRNPC, and WTAP, were screened to construct a nomogram model for predicting risk. The calibration curve, decision curve analysis, and clinical impact curve analysis were used to show that the nomogram model was effective and had a high net efficacy for risk prediction.
Conclusions:
m6A-related genes were correlated with immune cells. The nomogram model effectively predicted COVID-19 risk. Moreover, m6A-related genes may be associated with the presence or absence of symptoms in COVID-19 patients.
