Improving cancer driver genes identifying based on graph embedding hypergraph and hierarchical synergy dominance model

Zhipeng Hu, Xiaoyan Kui, Canwei Liu, Zanbo Sun, Shen Jiang, Min Zhang, Kai Zhu, Beiji Zou

Abstract: Highlights•Gene expression data and somatic mutation data are integrated to construct a global non-binary mutation matrix.•A new hypergraph construction method is developed using node2vec and K-means algorithms. Considering both the local topological features of genes and their exclusivity within each hyperedge, the entropy clustering coefficient (EntroCC) is introduced.•Gene expression data and DNA methylation data are integrated to enhance the differential expression scores obtained from gene expression data.•By analyzing the correlation among features, a multi-feature integration model, termed the hierarchical synergy dominance model (HSDM), is proposed.