Diarrhea is one of the major causes of mortality in young children worldwide, primarily within the third-world. Diarrheal disease can be caused by a variety of known viral (e.g. norovirus), bacterial (e.g. Vibrio cholerae, and parasitic (e.g. Giardia) pathogens, yet in many clinical cases the actual cause of the disease is unkown. Our study is aimed at providing a better understanding of the causes of diarrhea, with a goal of identifying potential new pathogens, or disruptions of the normal gastro-intestinal flora that lead to diarrhea.

We are relying on a large cohort of patients and matched controls comprising approximately 1000 children less than 18 months of age from Bangladeshy, Gambia, Kenya, and Mali, and are applying a variety of molecular methods (DGGE, 16S rRNA sequencing, microarrays) in order to characterize the microbial communities inhabiting their GI tract. The resulting data will be correlated with the results of microbiological, virological, and parasitological assays and with disease state.

This project has also led us to evaluate and develop methods for data analysis, in particular in the context of 16S rRNA sequencing. Existing software tools and analysis methodologies do not scale to the large numbers of sequences we are generating (currently our data comprises more than 1.5 million sequences). In addition, database searches, and other supervised clustering procedures, are limited in applicability due to the absence in public databases of sequences sufficiently related to the organisms found in gastro-intestinal samples.

Publications

James R. White, Niranjan Nagarajan, Mihai Pop. Statistical Methods for Detecting Differentially Abundant Features in Clinical Metagenomic Samples. PLoS Computational Biology, 5:e1000352 2009

Software

Metastats - statistical package for comparing clinical metagenomic data-sets.