Fig. 4: Validation of a diagnostic lipidomic signature in the validation cohort and the relationship with clinical features. | Nature Communications

Fig. 4: Validation of a diagnostic lipidomic signature in the validation cohort and the relationship with clinical features.

From: Identification and validation of a blood- based diagnostic lipidomic signature of pediatric inflammatory bowel disease

Fig. 4

a Plots depicting the log-transformed unit variance scaled distribution of LacCer(d18:1/16:0) and PC(18:1p/22:6) in individuals with IBD compared to symptomatic controls (SC) in the validation cohort (N = 117). The upper and lower lines of the boxes indicate the third and first quartiles, the lines in the middle represent the median, and the whiskers extending to the most extreme points within 1.5 times the IQR. Source data are provided as a Source Data file. b Receiver operating characteristic (ROC) curve illustrating the diagnostic prediction of pediatric inflammatory bowel disease (IBD) in the validation cohort using logistic regression. The model performance and validity measures were as follows: the area under the curve (AUC) for hsCRP was 0.73 (95% CI 0.63–0.82), while the AUC for the top two validated lipidomic markers, LacCer(d18:1/16:0) and PC(18:0p/22:6), was 0.87 (95% CI 0.80–0.94, P = 0.0004). Furthermore, the AUC for hsCRP in combination with the two top validated lipids was 0.87 (95% CI 0.80–0.94). c Pair-wise correlations of age, BMI, hsCRP, albumin, fecal calprotectin, LacCer(d18:1/16:0) and PC(18:0p/22:6) among all participants in the discovery cohort were assessed using Pearsons correlation coefficient (*P < 0.05, **P < 0.01 ***P < 0.001). BMI body mass index, hsCRP high-sensitivity C-reactive protein.

Back to article page