Skip to main content
Fig. 8 | Microbiome

Fig. 8

From: Microbial risk score for capturing microbial characteristics, integrating multi-omics data, and predicting disease risk

Fig. 8

Results for T1D prediction in the TEDDY study. A ROC curves and AUC values for predicting T1D status using various risk score. PRS_hla is constructed from the HLA alleles alone, and PRS is constructed from all SNPs found in the TEDDY cohort based on the existing PRS algorithm [49]. MRSα is the negative alpha diversity (Shannon index) calculated on the selected sub-community, which is selected by ANCOM-BC method and P+T method. B–D. Kaplan–Meier plots for the groups of subjects at high and low risk of developing T1D, based on PRS, MRSα, and the combination of PRS and MRSα, respectively. Subjects whose risk scores are above the third quartile are defined as high risk, others as low risk

Back to article page