SVM-based classification and feature selection methods for the analysis of Inflammatory Bowel disease microbiome data

Motivation: The human gut is one of the most densely populated microbial communities in the world. The interaction of microbes with human host cells is responsible for several disease conditions and of criticality to human health. It is imperative to understand the relationships between these microbial communities within the human gut and their roles in disease. Methods: In this study we analyze the microbial communities within the human gut and their role in inflammatory bowel disease (IBD). The bacterial communities were interrogated using Length Heterogeneity (LH-PCR) fingerprinting of mucosal and luminal associated microbial communities during healthy and diseases states. We develop support vector machine based classification and feature selection techniques to differentiate between healthy controls and patients suffering from IBD. Moreover, we develop site-specific classifiers to analyze community differences on the inner lining of the intestine (called mucosa) and the fluid within the intestine (called lumen).We also determine differentially abundant features across the different samples. Results: Using SVM-based classifiers with feature selection, we can distinguish the communities between the healthy controls and disease class patients. We also report differentially abundant features that exist between the different patient groups. The site-specific analysis provides an understanding of the microbial community differences between the lumen and mucosa of the healthy controls and patients suffering from IBD.