Fig. 5: Application of lipidIN to breast cancer clinical data.
From: LipidIN: a comprehensive repository for flash platform-independent annotation and reverse lipidomics
a Statistics of all annotations. b Results of Randomized Forest Important Indicators by Lipid subclass. c Receiver Operating Characteristic (ROC) of the first cohort as a training set and the second cohort as a test set. d Weighted correlation network analysis (WGCNA) of the first cohort with clinical indicators, the correlation was calculated with Pearson correlation coefficient, and cluster method was average hierarchical clustering. e, f Composition of several important modules of lipids in WGCNA of the first cohort.
