Background: The gut microbiota plays a crucial role in metabolic dysfunction-associated steatotic liver disease (MASLD). Next-generation sequencing technologies are essential for exploring the gut microbiome. While recent advancements in full-length 16S (FL16S) rRNA sequencing offer better taxonomic resolution, whether they establish stronger associations with the risk of MASLD remains to be determined.
Method: This study utilized long-read FL16S and short-read V3-V4 16S rRNA sequencing to profile gut microbiome compositions in age-, sex-, and BMI-matched case-control pairs of obese children with and without MASLD. A random forest predictive model was employed, using gut-microbiota features selected based on the top 35 most abundant taxa or a linear discriminant analysis score greater than 3. The model's performance was evaluated by comparing the area under the receiver operating characteristic curve (AUC) through a tenfold cross-validation method.
Results: Subjects with MASLD exhibited significantly elevated serum alanine aminotransferase, triglycerides, and homeostasis model assessment of insulin resistance levels compared to controls. At the genus level, the gut microbiome compositions detected by both FL16S and V3-V4 sequencing were similar, predominantly comprising Phocaeicola and Bacteroides, followed by Prevotella, Bifidobacterium, Parabacteroides, and Blautia. The AUC for the model based on FL16S sequencing data (86.98%) was significantly higher than that based on V3-V4 sequencing data (70.27%), as determined by DeLong's test (p = 0.008).
Conclusion: FL16S rRNA sequencing data demonstrates stronger associations with the risk of MASLD in obese children, highlighting its potential for real-world clinical applications.
Keywords: 16S rRNA; Gut microbiota; MASLD; Random forest; Sequencing.
© 2025. The Author(s).