Identification of a gene signature for discriminating metastatic from primary melanoma using a molecular interaction network approach.

ABSTRACT: Understanding the biological factors that are characteristic of metastasis in melanoma remains a key approach to improving treatment. In this study, we seek to identify a gene signature of metastatic melanoma. We configured a new network-based computational pipeline, combined with a machine learning method, to mine publicly available transcriptomic data from melanoma patient samples. Our method is unbiased and scans a genome-wide protein-protein interaction network using a novel formulation for network scoring. Using this, we identify the most influential, differentially expressed nodes in metastatic as compared to primary melanoma. We evaluated the shortlisted genes by a machine learning method to rank them by their discriminatory capacities. From this, we identified a panel of 6 genes, A

SUBMITTER: Metri R 

PROVIDER: S-EPMC5725601 | biostudies-literature | 2017 Dec

REPOSITORIES: biostudies-literature