Integrating metabolome and microbiome data – lessons for design, statistical analysis, and interpretation
The determination of the microbiome’s metabolic functions is a key challenge in understanding the contribution of the gut microbiome to health and disease. As metabolic functions are shared across phylogenetic classes, differences in microbiome composition do not necessarily translate in differences in metabolic output. Therefore, analyses of the microbiome composition alone cannot give conclusive insights into the collective metabolic output of a community. In the workshop, we discuss methodological challenges in the combined analysis of metabolome and microbiome data in the context of frequent life science research designs utilising observational data. The workshop will cover topics ranging from data normalisation over adequate statistical analysis to advanced modelling techniques for the integration of microbiome and metabolome data based on constraint-based modelling. The workshop will also cover important pitfalls in interpretation of multi-omics data. The learnt lessons will then be applied to data from Parkinson’s disease studies and colorectal cancer studies. Goal of the workshop is to enable the participants to make sound methodological choices when integrating microbiome and metabolome data, while understanding the limitations and potentials of the chosen methodology.
Indicate what the students should prepare beforehand and/or the level of knowledge required:
Experience with statistical analysis of omics data is of advantage, but is not a prerequisite. For preparation, read Hertel et al. (2021), Gut microbes (10.1080/19490976.2021.1915673)