Genomic Data: Quantitative Microbiome Profiling
- 1 minAs part of the Spring 2018 course in Genomic Data Manipulation (BST 281) taught by Curtis Huttenhower at Harvard Chan, we had discussed quantitative analytical issues that arose in the most recent ‘omics publications in the style of a journal club. Specifically, my group reviewed and presented to the class a late-2017 Nature paper that proposed quantitative microbiome profiling (QMP) to address the issue of microbial load variability across samples, which has been a major limitation of relative microbiome profiling (RMP) - the current most common aproach to measuring microbial composition in the human gut. Thus, for our final group project, we applied our own variant of the QMP method on longitudinal fecal samples from a participant with Crohn’s Disease (CD) vs. a healthy control from the NIH Human Microbiome Project (HMP2).
- Our journal club presentation displayed above summarizes the QMP paper.
- The final project presentation below gives a quick overview/analysis plan with preliminary results that generated class discussions and feedback that were then taken into account as we finished up the project with our individual write-ups.
- Our project’s Github repo contains all the data manipulation and analysis programs written in python, R, and mathematica.
- You can read my final individual write-up for the project: here.
Eunice Yeh
Harvard Biostatistician