Page 15 - MDC Abstract Book & Guide

Abstract
Clinical Microbiomics is a contract research organization (CRO) that offers custom-tailored microbiome analysis for clinical, pre-clinical and animal health studies. The service includes DNA extraction, sequencing, profiling, and biostatistics linking the microbiome to clinical data and outcomes. As a science driven CRO, the focus lays on the continuous development of novel ways of analyzing the microbiome. Ultra-high resolution microbiomics allows for identification of single-nucleotide variance (SNV) between sample-specific populations (strains) of the same bacterial species and thus discriminates between sample specific populations found in different subjects. Moreover, this resolution accounts for differences in functional potential and gene content. Here, we demonstrate how a tree- based phylogenetic analysis facilitates powerful statistical analysis and identification of functional links through ancestral reconstruction.
Bill Shannon BioRankings
Bill Shannon has over 25 years of experience developing analytical methodology for complex and big data in biomedical R&D. In his capacity as a Professor of Biostatistics in Medicine at Washington University School of Medicine in St. Louis, Bill focused on statistical methods grants in emerging areas of medicine, and worked with researchers when there are no known statistical methods for analyzing their study data. He retired in 2017 as Professor Emeritus and Co-Founded and is Managing Partner of BioRankings providing multi-omics and other statistical analysis to medical and agricultural companies. Bill acts as a partner with clients and works to communicate results and their implications to both business executives and scientific researchers. Dr. Shannon earned his PhD in Biostatistics at the University of Pittsburgh in 1995, and completed his MBA at Washington University in St. Louis Olin School of Business in 2012.
Presentation Title
New Statistical Method Identifies Cytokines that Distinguish Stool Microbiomes
Abstract
A wide class of formal statistical methods exist that can be developed and applied to microbiome data. In this talk we present a statistical regression model. Regressing an outcome or dependent variable onto a set of input or independent variables allows the analyst to measure associations between the two so that changes in the outcome can be described by and predicted by changes in the inputs. While there are many ways of doing this in classical statistics, where the dependent variable has certain properties (e.g., a scalar, survival time, count), little progress on regression where the dependent variable are microbiome taxa counts has been made that do not impose extremely strict conditions on the data. In this paper, we propose and apply a new regression model combining the Dirichlet- multinomial distribution with recursive partitioning providing a fully non-parametric regression model. This model, called DM-RPart, is applied to cytokine data and microbiome taxa count data and is applicable to any microbiome taxa count/metadata, is automatically fit, and intuitively interpretable.