A systems biology-based classifier for hepatocellular carcinoma diagnosis

Area of Science:

• Biomedical Informatics
• Systems Biology
• Oncology

Background:

• Early diagnosis of hepatocellular carcinoma (HCC) is critical for effective treatment and improved patient survival.
• Existing gene expression-based HCC markers lack consistency and validation.
• A novel approach is needed to enhance the accuracy of HCC diagnosis.

Purpose of the Study:

• To develop a systems biology-based classifier for improved early-stage hepatocellular carcinoma (HCC) diagnosis.
• To integrate differential gene expression with protein interaction network topology for enhanced diagnostic performance.

Main Methods:

• Utilized Oncomine platform to identify differentially expressed genes in HCC tissues.
• Analyzed gene networks using GeneGO Meta-Core software to identify hub genes.
• Constructed a Partial Least Squares model using hub gene expression data for HCC classification.

Main Results:

• The developed classifier demonstrated high predictive accuracy (85.88–92.71%) and an area under the ROC curve close to 1.0.
• Integration of network topological features significantly improved predictive performance.
• The modeling strategy showed applicability to HCC and other cancer types.

Conclusions:

• A systems biology-based classifier integrating gene expression and protein network features enhances HCC diagnostic performance.
• This approach offers a promising strategy for improving cancer diagnostics.