Bile acid profiling as an effective biomarker for staging in pediatric inflammatory bowel disease.

Rapid and accurate clinical staging of pediatric patients with inflammatory bowel disease (IBD) is crucial to determine the appropriate therapeutic approach. This study aimed to identify effective, convenient biomarkers for staging IBD in pediatric patients. We recruited cohorts of pediatric patients with varying severities of IBD to compare the features of the intestinal microbiota and metabolites between the active and remitting disease stages. Metabolites with potential for staging were targeted for further assessment in both patients and colitis model mice. The performance of these markers was determined using machine learning and was validated in a separate patient cohort. Pediatric patients with IBD exhibited distinct gut microbiota structures at different stages of disease activity. The enterotypes of patients with remitting and active disease were Bacteroides-dominant and Escherichia-Shigella-dominant, respectively. The bile secretion pathway showed the most significant differences between the two stages. Fecal and serum bile acid (BA) levels were strongly related to disease activity in both children and mice. The ratio of primary BAs to secondary BAs in serum was developed as a novel comprehensive index, showing excellent diagnostic performance in stratifying IBD activity (0.84 area under the receiver operating characteristic curve in the primary cohort; 77% accuracy in the validation cohort). In conclusion, we report profound insights into the interactions between the gut microbiota and metabolites in pediatric IBD. Serum BAs have potential as biomarkers for classifying disease activity, and may facilitate the personalization of treatment for IBD.

Multi-view clustering via multi-manifold regularized non-negative matrix factorization.

Non-negative matrix factorization based multi-view clustering algorithms have shown their competitiveness among different multi-view clustering algorithms. However, non-negative matrix factorization fails to preserve the locally geometrical structure of the data space. In this paper, we propose a multi-manifold regularized non-negative matrix factorization framework (MMNMF) which can preserve the locally geometrical structure of the manifolds for multi-view clustering. MMNMF incorporates consensus manifold and consensus coefficient matrix with multi-manifold regularization to preserve the locally geometrical structure of the multi-view data space. We use two methods to construct the consensus manifold and two methods to find the consensus coefficient matrix, which leads to four instances of the framework. Experimental results show that the proposed algorithms outperform existing non-negative matrix factorization based algorithms for multi-view clustering.