Additional file 1: of Multi-omics differentially classify disease state and treatment outcome in pediatric Crohn’s disease

Figure S1. Stacked bar-chart showing the relative abundance of microbial classes across the metagenomic sequencing data. Figure S2. Stacked bar-chart showing the relative abundance of bacterial classes across the 16S rRNA gene sequencing data. Figure S3. Barplots comparing random forest model accuracies based on non-rarefied-centered log-ratio transformed and rarefied 16S rRNA gene taxa abundances. Figure S4. Boxplots of genetic risk scores and the number of observed OTUs, between Crohn’s disease and control patients. Figure S5. Distribution of variable importance for 16S rRNA gene-identified genera disease classification. Figure S6. Boxplots of the relative abundance of the top 16S rRNA gene sequencing-identified genera for classifying disease state compared between sequencing technologies. Figure S7. Boxplots of genetic risk score and alpha-diversity between responders and non-responders to treatment. Figure S8. Pairwise Spearman correlation coefficients between features in the combined disease random forest. Figure S9. Pairwise Spearman correlation coefficients between features in the combined treatment response random forest. Figure S10. Features ranked by their relative importance for classifying disease state in the RISK validation cohort. Table S1. Table of demographic and phenotypic characteristics of pediatric patients. Table 2. Table of phenotypic characteristics and treatments of pediatric patients. (PDF 2350 kb)