Fig. 2From: Molecular signature comprising 11 platelet-genes enables accurate blood-based diagnosis of NSCLCSchematic representation of workflow and discovery of gene-signature. (a) The upper panel is a schematic representation illustrating the underlying methodology implemented for the identification of the concise gene-panel utilizing RNA-seq data of Tumor Educated Platelets (TEPs) (GSE68086). The lower panel represents the experimental design and the downstream statistical analysis employed in the validation of the inferred signature on a geographically distinct NSCLC patient cohort. (b) A comparison between different feature selection methods shows that a combination of Coefficient of Variation (CV) and Analysis of Variance (ANOVA) performs the best. (c) Classification accuracy across different cancer typesBack to article page