Skip to main content
Fig. 1 | BMC Bioinformatics

Fig. 1

From: Interpretable prediction of necrotizing enterocolitis from machine learning analysis of premature infant stool microbiota

Fig. 1

Collection timing and microbiota characterization of neonatal stool samples from the two cohorts used for NEC prediction. Warner et al. [31] and Olm et al. [27] prospectively collected stool samples for subsequent sequence-based microbiota analysis from hospitalized preterm infants at variable timepoints throughout the first several months of life. A subset of enrolled infants eventually developed NEC. For this study, 2895 samples from Warner et al. and 700 from Olm et al. were used for development and retrospective application of a NEC prediction model (A, B). Kraken2 was used to perform microbiota member classifications for all available samples from both studies. Aligned bacterial classifications at all taxonomic levels are shown for both studies (C)

Back to article page