MONet: cancer driver gene identification algorithm based on integrated analysis of multi-omics data and network...

Area of Science:

• Oncology
• Bioinformatics
• Computational Biology

Background:

• Cancer progression is driven by mutations in specific genes, known as driver genes.
• Identifying these driver genes is critical for developing targeted therapies and biomarkers.
• Current methods may miss crucial driver genes due to data complexity.

Purpose of the Study:

• To introduce MONet, a novel algorithm for identifying cancer driver genes.
• To leverage multi-omics data and network analysis for enhanced driver gene detection.
• To discover previously unidentified cancer driver genes.

Main Methods:

• Integrated analysis of multi-omics data and network models.
• Utilized two graph neural network algorithms on protein-protein interaction (PPI) networks.
• Employed a multi-layer perceptron (MLP) model for semi-supervised driver gene identification.

Main Results:

• MONet demonstrated robustness across diverse PPI networks and outperformed baseline models.
• Achieved superior performance in area under the receiver operating characteristic curve and precision-recall curve.
• Identified 37 novel cancer driver genes, including 29 validated in existing literature (e.g., APOBEC2, GDNF, PRELP).

Conclusions:

• MONet successfully identifies known and novel cancer driver genes.
• The algorithm provides biologically meaningful insights into cancer mechanisms.
• MONet enhances the discovery of potential therapeutic targets and biomarkers.