Determining the edge weight between ZFP106 and TRIM9 genes. Alexander Karsakov Thomas Bartlett Artem Ryblov Iosif Meyerov Mikhail Ivanchenko Alexey Zaikin 10.1371/journal.pone.0169661.g002 https://plos.figshare.com/articles/figure/Determining_the_edge_weight_between_ZFP106_and_TRIM9_genes_/4574272 <p>Each point corresponds to gene methylation levels in the control group (green), other BRCA-negative (blue) and BRCA-positive (red) subjects. The solid line shows the linear regression <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0169661#pone.0169661.e001" target="_blank">model (1)</a>. While the data sets for healthy and tumour samples are quite distinct, the mismatch <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0169661#pone.0169661.e002" target="_blank">Eq (2)</a> is of the same order of magnitude for both classes. Employing the Mahalanobis distance <a href="http://www.plosone.org/article/info:doi/10.1371/journal.pone.0169661#pone.0169661.e003" target="_blank">Eq (3)</a> overcomes this problem.</p> 2017-01-20 17:33:03 classification 15295 gene methylation levels Parenclitic Network Analysis control group subjects DNA methylation data 12 network topology indices power-law node degree distribution parenclictic networks