Integration of Multi-Omics Data With Topology Adaptive Graph Convolutional Network for Cancer Driver Gene...

Area of Science:

• Computational biology
• Genomics
• Bioinformatics

Background:

• High-throughput molecular data analysis presents computational challenges for identifying cancer driver genes.
• Existing methods struggle with the complexity and volume of multi-omics data.

Purpose of the Study:

• To introduce ATTAG, a novel framework for predicting cancer driver genes.
• To integrate multi-omics data with biomolecular networks using topology-adaptive graph neural networks.

Main Methods:

• ATTAG constructs three gene networks: protein interactions, semantic similarities, and pathway co-occurrence.
• Graph convolutional networks (GCNs) generate embeddings from networks and multi-omics data.
• Topology-adaptive graph neural networks refine embeddings for driver gene prediction in Bladder Cancer (BLCA).

Main Results:

• ATTAG demonstrates exceptional performance in identifying cancer driver genes in BLCA.
• The framework accurately identifies both known and novel candidate cancer genes.
• ATTAG outperforms current state-of-the-art methods.

Conclusions:

• ATTAG provides an effective computational framework for cancer driver gene identification.
• The integration of multi-omics data and biomolecular networks enhances prediction accuracy.
• ATTAG has the potential to advance cancer research and therapeutic target discovery.