biostudies-literatureMetri RCancer Research UKWorldwide Cancer Research17314https://www.ebi.ac.uk/biostudies/studies/S-EPMC5725601biostudies-literatureUnknown7(1)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, AScientific reportsIdentification of a gene signature for discriminating metastatic from primary melanoma using a molecular interaction network approach.PMC5725601C588/A19167C8216/A61291916712-00231058911963Metri RChandra NMohan AMolina-Paris CBishop DNewton-Bishop JPozniak JNsengimana JfalseIdentification of a gene signature for discriminating metastatic from primary melanoma using a molecular interaction network approach.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, A2017-01-01T00:00:00Z2017 Dec2025-06-01T00:36:05.464Z2025-06-01T00:36:05.464ZS-EPMC57256012922993610.1038/s41598-017-17330-0