Extracting Insights On The Dynamic Health-Disease Transitions In The Human Gut Microbiome

Smarr, Larry, Marc Jaffrey, Michael Dushkoff, Brynn Taylor, Pilar Ackerman, Mehrdad Yezdani, and Weizhong Li. “Extracting Insights On The Dynamic Health-Disease Transitions In The Human Gut Microbiome” (2018).

Abstract
The trillions of microbes living in our large intestine — the gut microbiome — play a profound role in human health and disease [1]. While much has been done to explore its diversity, a full understanding of how the dynamical evolution of the microbiome ecology influences healthy and disease states is only beginning to be understood [2].

In this article, we start by reviewing previous research results, [3] examining the gut microbiome taxonomic differences between healthy people and of those suffering from the three subtypes of the autoimmune Inflammatory Bowel Disease (IBD): Ileal Crohn’s, Colonic Crohn’s, and Ulcerative Colitis [6]. This study was recently expanded to understand the functional differences by using the Kyoto Encyclopedia of Genes and Genomes (KEGG) [4] protein families in the gut microbiome of the samples. Each of the entries in the KEGG database describes an orthologous protein family, which have specific biological functions. In the study relative abundances of 10,192 KEGG protein families were computed from the sequencing of the human stool samples. Using traditional machine learning techniques, it was shown [5] that a subset of the KEGG protein families can distinguish between healthy and the IBD states. In this paper, we describe the results obtained with the Pattern Computer proprietary algorithms, tools, and techniques, using an approach without prior assumptions on this large dataset of 62 human microbiome samples, each with the relative abundance of the ~10,000 KEGG protein families. We identified 39 KEGG protein families that were significant in differentiating the disease states from each other and from healthy states, with 9 of the KEGG protein families (out of ~10,000 total) being most associated with a dynamic path from disease to health in the human-gut microbiome. With the Pattern Computer approach, we reduced the size of the dataset to be analyzed by three orders of magnitude. The biochemical pathways, that 6 out of the 9 KEGG protein families are associated with, suggest a hypothesis for further study: Inflammatory bowel disease (IBD), like other inflammatory diseases, may be associated with abnormal oxidative phosphorylation or oxidative stress.

LINK TO ARTICLE