Fig. 4 | npj Systems Biology and Applications

Fig. 4

From: Integrating personalized gene expression profiles into predictive disease-associated gene pools

Fig. 4

Integrating the personalized profiles into a predictive pool of disease-associated genes. a, b Distribution of the number of individual perturbation profiles in which a gene appears for a control and b case subjects of the three considered diseases. The two curves in each panel correspond to the actual data and the random expectation according to a model of randomly selected genes (green). c Venn diagram of three broad gene pools compiled from genes that are in at least X individual perturbation profiles. d The ROC for the disease state classification by the fraction of the broad gene pool that is contained in a subject’s perturbation profile. The AUC values are 0.77 ± 0.03, 0.81 ± 0.06 and 1.0 ± 0.0 for asthma, PD and HD, respectively (mean value ± standard deviation computed over 100 cross-validations). e Sensitivity and specificity as a function of the fraction of broad gene pool for asthma (z thresh = 2.5; X = 10). f Illustration of the disease model suggested by the analysis of PEEPs

Back to article page