Serum micro-RNAs with mutation-targeted RNA modification: a potent cancer detection tool constructed using an...

Area of Science:

• Molecular Biology
• Genomics
• Cancer Research

Background:

• RNA modifications are crucial in biological processes and disease, with micro-RNAs (miRNAs) playing a key role.
• Genetic variants targeting RNA modifications in miRNAs can significantly alter their function and impact cancer development.

Purpose of the Study:

• To investigate the association between RNA modification-related miRNA variants (RMvar-related miRNAs) and tumorigenesis.
• To develop a cancer-specific diagnostic signature using RMvar-related miRNAs and machine learning.

Main Methods:

• Utilized an optimized machine learning workflow combining LASSO, RFE, and nine ML algorithms.
• Screened 504 serum RMvar-related miRNAs to identify candidate biomarkers.
• Constructed and validated a diagnostic signature using a large cohort of 43,047 clinical samples.

Main Results:

• The developed signature demonstrated exceptional diagnostic performance: AUC of 0.998, 93.1% specificity, and 99.3% sensitivity in the validation cohort.
• The signature showed satisfactory diagnostic capabilities across various cancers and conditions.
• Successfully distinguished early-stage tumors, highlighting its potential for early cancer detection.

Conclusions:

• RMvar-related miRNAs are closely associated with tumor development.
• The proposed miRNA-based signature serves as an effective tool for cancer screening and early detection.