Related Experiment Video
Updated: Mar 19, 2026
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Construction and Validation of an N7-Methylguanosine-Related Prognostic Model for Acute Myeloid Leukemia
Lina Wang1, Ming Li1, Yaming Xi1
1Department of Hematology, The First Hospital of Lanzhou University, Lanzhou, Gansu, People's Republic of China.
Purpose:
Increasing evidence suggests the involvement of N7-methylguanosine (m7G) in cancer biology. However, its role in acute myeloid leukemia (AML) remains unclear. Herein, bioinformatics approaches were used to obtain insights for AML risk stratification and treatment.
Patients And Methods:
Data from TCGA-LAML, GSE114868, and GSE37642 were analyzed. Differentially expressed genes from GSE114868 were intersected with key module genes identified via weighted gene co-expression network analysis. The identified genes underwent univariate and multivariate Cox regression analyses and machine learning to identify prognostically relevant m7G-related genes (m7G-RGs). Moreover, a prognostic risk model was built and validated, and its association with immune infiltration was evaluated. Model performance was compared with the European LeukemiaNet (ELN) 2022 genetic risk stratification system. Expression levels of key genes were analyzed in GSE114868 and validated in independent clinical samples.
Results:
A prognostic risk model was developed based on seven m7G-RGs (TM6SF1, IL1R2, MTX1, SUSD3, SLC22A4, TUBA4A, and RETN). Patients with AML were stratified into high- and low-risk groups, with the low-risk group showing significantly longer overall survival. Compared with the ELN 2022 classification, the model provided significant prognostic refinement within the heterogeneous intermediate-risk subgroup. IL1R2 and TUBA4A expression were significantly associated with immune cell infiltration scores. Consistent with the bioinformatics analyses, TM6SF1, IL1R2, MTX1, and SLC22A4 expression was significantly reduced in an independent AML cohort and in patient samples compared with controls.
Conclusion:
We developed and validated a novel m7G-related prognostic model for AML based on seven genes. The findings suggest a potential association among m7G modification, AML prognosis, and the tumor immune microenvironment. The model showed complementary value to the ELN 2022 risk stratification by improving risk assessment among patients classified as intermediate risk. As a retrospective, computational study, prospective validation is required. The identified m7G-RGs warrant further investigation as potential biomarkers or therapeutic targets.
