Identification of Novel Potential Genes Involved in Cancer by Integrated Comparative Analyses

Abstract

The main hallmarks of cancer diseases are the evasion of programmed cell death, uncontrolled cell division, and the ability to invade adjacent tissues. The explosion of omics technologies offers challenging opportunities to identify molecular agents and processes that may play relevant roles in cancer. They can support comparative investigations, in one or multiple experiments, exploiting evidence from one or multiple species. Here, we analyzed gene expression data from induction of programmed cell death and stress response in Homo sapiens and compared the results with Saccharomyces cerevisiae gene expression during the response to cell death. The aim was to identify conserved candidate genes associated with Homo sapiens cell death, favored by crosslinks based on orthology relationships between the two species. We identified differentially-expressed genes, pathways that are significantly dysregulated across treatments, and characterized genes among those involved in induced cell death. We investigated on co-expression patterns and identified novel genes that were not expected to be associated with death pathways, that have a conserved pattern of expression between the two species. Finally, we analyzed the resulting list by HumanNet and identified new genes predicted to be involved in cancer. The data integration and the comparative approach between distantly-related reference species that were here exploited pave the way to novel discoveries in cancer therapy and also contribute to detect conserved genes potentially involved in programmed cell death.

Computational Approaches for Cancer-Fighting: From Gene Expression to Functional Foods

“…We here present a deeper investigation on the 734 genes from [26]. These genes are co-expressed in 9 apoptotic treatments (Tamoxifen (TMX) [32], Zika virus (NCC-Zika and HNP-Zika) [33], Vitamin C (Vitamin C) [34], ethanol (Ethanol) [35], a CBFβ-SMMHC inhibitor (Al-10-49) [36], Pam3CSK4 (PAM3h and PAM24h) [37] and a KDM1A inhibitor (NCD38) [38]), together with genes involved in programmed cell death.…”

“…Among these, we considered the nutrigenomics treatments that are known to induce apoptosis [34][35][36][37][38][39][40][41][42][43][44][45][46][47][48], and considered those genes with a False Discovery Ratio (FDR) < 0.05 and a |log2 Fold Change| > 1.5. The comparison of the gene expression patterns in apoptosis determining nutrigenomics treatments with those of the 9 independent apoptotic treatments from [26] resulted in a list of 149 differentially expressed genes showing the same trend (up-or down-regulation) in the two classes of apoptotic treatments.…”

“…The 734 genes that resulted co-expressed with genes involved in programmed cell death in 9 apoptotic treatments [26] were investigated exploiting NutriGenomeDB [25]. This analysis resulted in 149 genes that showed the same expression pattern in 15 nutrigenomics experiments inducing apoptosis (Table S1).…”

“…We recently identified 734 genes that resulted co-expressed with genes involved in programmed cell death in 9 apoptotic treatments [26]. They were not yet reported as involved in apoptosis by GO and/or KEGG.…”

“…Based on the appropriate exploitation of currently available bioinformatics resources, we recently identified 734 genes that resulted co-expressed with genes involved in programmed cell death but not reported as implicated in this process by gene ontology (GO) or Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways [26]. We further investigated this collection with the principal aim to identify those genes that have the same expression pattern between apoptotic previous results and nutrigenomics treatments inducing cell death [25].…”