2014-15: Bioinformatics: Discovering Patterns in Human Microbiome Data (HMD): March 16-18, 2015


This workshop was held at SAMSI in Research Triangle Park, N.C.


The human microbiome is the community of microorganisms that reside in different body habitats. Host-microbiome interactions play an important but poorly-characterized role in health. Advances in high-throughput technologies including next-generation sequencing and mass spectroscopy allow us to query the population of micro-organisms and observe the interactions in greater detail. The effect of choices in measurement protocols, the multivariate longitudinal nature of data, and combining measurements on different scales pose difficulties in discovering and confirming significant relationships in the human microbiome.

The aim of this workshop was to create a forum for ideas for overcoming current and future challenges in the analysis of human microbiome data. In this workshop, participants learned how metagenomic (sequence-based) and metabolomic (mass spectroscopy-based) data are generated and the implications for analysis. Gaps in the current state-of-the-art methods were highlighted, particularly with respect to the analysis of multivariate longitudinal data and the use of statistical experimental design to assess bias. Participants assembled into research groups and were asked to formulate plans to address the deficiencies of current methods.

Participants could optionally submit an abstract for a short presentation or poster.  A small subset of the submitted abstracts were selected for short presentations at the workshop.

Questions: email bioinformatics@samsi.info

Schedule and Supporting Media

Participant List
Speaker Titles and Abstracts

Monday, March 16, 2015

Time Description Speaker Slides Videos
8:00-8:30 Registration
8:30-8:45 Opening Remarks Snehalata Huzurbazar, University of Wyoming
Paul Brooks, Virginia Commonwealth University
8:45-9:15 SAMSI Working Group Summary and Workshop Introduction
9:15-10:15 Generalizing PCA to Accommodate for the Multiple Sources of Data and Constraints in the Human Microbiome Susan Holmes, Stanford University
10:15-10:30 Break
10:30-11:30 Predictive Models of Microbiome Dynamics: Designing Bacterial Cocktails to Ameliorate Enteric Infection and to Stimulate Immune Systems Vanni Bucci, University of Massachusetts Dartmouth pdf
11:30-12:00 Breakout Meeting Discussion
12:00-1:30 Lunch
1:30-3:30 Breakout Research Meetings
Groundtruthing 16S experiments – Room 150
Interaction inference in longitudinal data – Room 203
Predicting community state type changes – Room 259
Sparse multiway data integration – Room 219
3:30-4:00 Breakout Meeting Reports
4:00-5:30 Short Talk Session
Comparing Clostridium Difficile Infected Patients before and after a Treatment Using Loops in DNA Sequences Giseon Heo, University of Alberta
Alpha Diversity of the Vaginal Microbiome Clusters within Families: a Twin Study Michelle Wright, Virginia Commonwealth University
Bayesian Variable Selection for Multinomial-Dirichlet Regression with an Application to Microbiome Data Integration Duncan Wadsworth, Rice University
Vaginal Microbiome: Comparison of Bacterial Vaginosis 16S rRNA Taxa Composition Ekaterina Smirnova, University of Wyoming
5:30-7:30 Poster Session and Reception

Tuesday, March 17, 2015

Time Description Speaker Slides Videos
8:30-8:45 Registration and Announcements Paul Brooks, Virginia Commonwealth University
9:00-10:00 Statistical and Visualization Methods for Metagenomic Analysis Hector Corrada Brava, University of Maryland
10:00-12:00 Breakout Research Meetings
Predicting community state type changes – Room 150
Sparse multiway data integration – Room 259
Additional groups – Room 219
12:00-1:30 Lunch
1:30-2:00 Breakout Meeting Reports
2:00-3:00 Shanaka Wijesinghe, Virginia Commonwealth University
3:00-3:30 Break
3:30-5:00 Short Talk Session
Analysis of Longitudinal Microbiota Data Brandie D. Wagner, University of Colorado pdf
High-Sensitivity Pattern Discovery in High-Dimensional Heterogeneous Datasets Gholamali Rahnavard, Harvard School of Public Health pdf
Multivariate Association of Microbial Communities with Rich Metadata in High-Dimensional Studies Ayshwarya Subramanian, Harvard School of Public Health

Wednesday, March 18, 2015

Time Description Speaker Slides Videos
8:45-9:00 Registration and Announcements: Paul Brooks, Virginia Commonwealth University
9:00-10:00 Microbiome Power/Sample Size Calculations (plus a bit of formal hypothesis testing) Bill Shannon, Washington University in St. Louis
10:00-12:00 Breakout Research Meetings
Interaction inference in longitudinal data – Room 150
Analyzing human microbiome data – Room 259
Additional groups – Room 219
12:00-1:00 Breakout Meeting Reports and Discussion
1:00 Adjourn and Box Lunch