作者: Alessandra J. Conforte , Jack Adam Tuszynski , Fabricio Alves Barbosa da Silva , Nicolas Carels
关键词:
摘要: Traditional approaches to cancer therapy seek common molecular targets in tumors from different patients. However, profiles differ between patients, and most exhibit inherent heterogeneity. Hence, imprecise targeting commonly results side effects, reduced efficacy, drug resistance. By contrast, personalized medicine aims establish a diagnosis specific each patient, which is currently feasible due the progress achieved with high-throughput technologies. In this report, we explored data human RNA-seq protein-protein interaction (PPI) networks using bioinformatics investigate relationship tumor entropy aggressiveness. To compare PPI subnetworks of sizes, calculated Shannon associated vertex connections differentially expressed genes comparing samples their paired control tissues. We found that inhibition up-regulated connectivity hubs led higher reduction subnetwork compared obtained selected at random. Furthermore, these were described be participating processes. also significant negative correlation entropies respective 5-year survival rates corresponding types. This was observed considering patients lung squamous cell carcinoma (LUSC) adenocarcinoma (LUAD) based on clinical The Cancer Genome Atlas database (TCGA). Thus, network increases parallel aggressiveness but does not correlate size. consistent previous reports allowed us assess number inhibited for effective, context precision medicine, by reference 100% patient rate 5 years after diagnosis. Large standard deviations variations target numbers per among types characterize