Abstract: Bioinformatics analysis approaches were employed to screen the differential genes from the genomic and epigenetic data of the patients with T-cell acute lymphoblastic leukemia (T-ALL), and the multi-omics gene similarity fusion networks were built with a view to screen out key genes and explore its pathogenic mechanisms. The data of RNA-seq, CTCF ChIP-seq and DNA methylation of T-ALL were downloaded from GEO and SRA databases. Using both DESeq2 and edgeR software, the differential gene expression analysis of RNA-seq and CTCF ChIP-seq data was performed. The CHAMP software was adopted to screen the differential genes in DNA methylation data. Since then, 5 887, 5 315 and 2 196 differential genes had been identified from the data of RNA-seq, CTCF ChIP-seq and DNA methylation, respectively. There were 119 genes in the intersection of the three differential gene sets. The multi-omics gene similarity fusion network was constructed, and 48 key genes with strong interactions and more associations were screened out from it. Gene Ontology (GO) and KEGG path enrichment analysis were performed for the 48 key genes, the protein-protein interaction network of the key genes was established by using the STRING database, Cytoscape software was used to select eight core genes (CTLA4, CD7, GPR29, CD5, CD247, IL2RB, FASLG and CD274). After comprehensive searches in CGC and CTD databases, the results indicate that the eight core genes hold great potential of becoming the T-ALL's biomarkers, and they provide assistance to the exploration of pathogenesis of T-ALL and targeting drug development.

Key words: T-ALL, similarity fusion network, differential genes, key genes, bioinformatics