Subscribe to RSS
DOI: 10.1055/s-0037-1605007
Superiority of multiple random gene sets in predicting survival of patients with hepatocellular carcinoma
Publication History
Publication Date:
02 August 2017 (online)
Despite multiple publications, molecular signatures predicting the course of hepatocellular carcinoma (HCC) have not yet been integrated into clinical routine decision making. Given the number and diversity of signatures published throughout the past decade, optimal number, best combinations, and benefit of functional associations of genes in prognostic signatures still remain to be defined. We therefore investigated a vast number of randomly chosen gene sets in order to encompass the full range of prognostic gene sets.
Depending on the selected size, 4.7 to 23.5% of all random gene sets exhibit prognostic potential by separating patient subgroups with significantly diverse survival. This finding was further substantiated by investigating predefined gene sets and signaling pathways from KEGG, Biocarta, Reactome, PID, and GO terms, also resulting in a comparable high number of significantly prognostic gene sets.
However, combining multiple random gene sets using „swarm intelligence” resulted in a significantly improved predictability of prognosis for approximately 63% of all patients. In these patients, approx. 70% of all random 50-gene containing gene sets resulted in equal and stable prediction of survival. For all other patients a reliable prediction seems highly unlikely for any selected gene set. Using a machine learning and independent validation approach, we demonstrated a high reliability of random gene sets and swarm intelligence in HCC prognosis.
In conclusion, our data raise concerns on the feasibility of establishing a single robust transcriptome based gene set to predict HCC prognosis. However, we demonstrate that simultaneous use of multiple gene sets for prognosis prediction may not only be superior but also more robust for predictive purposes. These more robust and quality estimated gene sets using swarm intelligence may pave the way to further prospective studies and ultimately to clinical routine application using „swarm intelligence” for prognostic evaluation of HCC.