CBNA

CBNA (Controllability based Biological Network Analysis) is a computational framework that identifies coding and non-coding cancer driver genes, including miRNA drivers. The framework integrates various genomic data types, such as gene expression, gene networks, and mutation data, to comprehensively analyze cancer genomics.

The CBNA framework consists of two main stages:

1. Building a network for a specific condition, such as cancer.

2. Identifying driver genes based on the constructed network.

When applied to the BRCA dataset, CBNA demonstrated superior performance to existing methods in detecting coding cancer drivers. Additionally, it predicted 17 miRNA drivers for breast cancer, some of which have been validated by literature, while others serve as promising candidates for wet lab validation.

CBNA can also detect subtype-specific cancer drivers, and several of its predicted drivers have been confirmed to be related to specific breast cancer subtypes. Furthermore, the framework can be applied to discover drivers of epithelial-mesenchymal transition (EMT), with seven coding and six miRNA drivers predicted by CBNA in known EMT gene lists.