A novel prior knowledge embedded dynamic multi-objective optimization algorithm termed TRDMO-P significantly improves the detection accuracy of early warning signals for disease critical biomarkers compared with state-of-the-art methods, according to a methods study published in Frontiers in Bioinformatics.
The authors state that designing optimization methods for identifying critical biomarkers is essential for early clinical intervention within cancer precision medicine. However, existing optimization methods demonstrate limited performance because they rely on static network analysis. By depending on static networks, previous approaches ignore the temporal dynamics of gene interaction networks and lack the effective integration of biological prior knowledge during the optimization process.
To address these performance limitations, the research team developed TRDMO-P for disease critical biomarker identification.
The TRDMO-P algorithm operates through two core strategies designed to exploit dynamic network characteristics. The first core strategy is prior biological gene-guided population initialization, which is designed to enable an effective optimization search and achieve faster algorithmic convergence. The second core strategy is network topology degree-guided evolution, which functions to retain effective biological genes and ensure biologically consistent optimization.
The researchers tested TRDMO-P in experiments utilizing cancer datasets from The Cancer Genome Atlas as well as esophageal single-cell transcriptome data. In these evaluations, TRDMO-P demonstrated significant improvements in the detection accuracy of early warning signals when compared directly against state-of-the-art methods. In addition to benchmark testing, the authors conducted ablation studies that validated the specific effectiveness of both proposed strategies within the algorithm.
Based on their experimental findings and ablation results, the authors conclude that TRDMO-P enables reliable and biologically meaningful identification of critical biomarkers. The research team notes that by capturing temporal dynamics and integrating biological prior knowledge, the proposed dynamic network biomarker identification method shows strong potential for future applications in cancer precision medicine.
Subject of Research: Dynamic multi-objective optimization for cancer biomarker identification
Article Title: Prior knowledge embedded dynamic multi-objective optimization algorithm for identifying cancer critical biomarkers
Article References: Cheng, P., Zhao, M., Jia, G., Ren, X., Ma, C., & Guo, W.-F. (2026). Prior knowledge embedded dynamic multi-objective optimization algorithm for identifying cancer critical biomarkers. Frontiers in Bioinformatics, 6, 1950833. https://doi.org/10.3389/fbinf.2026.1950833
Image Credits: AI Generated
DOI: 10.3389/fbinf.2026.1950833
Keywords: TRDMO-P, cancer precision medicine, critical biomarkers, dynamic multi-objective optimization, gene interaction networks, temporal dynamics, biological prior knowledge, early clinical intervention, population initialization, network topology degree, biologically consistent optimization, early warning signals, TCGA cancer datasets, esophageal single-cell transcriptome, dynamic network biomarkers, evolutionary computation, ablation studies
News Source: Nathaniel Bowman. (November 17, 2026). Dynamic Multi-Objective Optimization Algorithm Improves Detection of Cancer Critical Biomarkers. Scienmag.



