ADMIRE: Association, causality and biomarker Discovery in translational MIcrobiome REsearch. (R-11406)

The introduction of next generation sequencing technologies has pushed microbiome research to previously unseen levels, opening a route towards improved sustainable crop production and disease prevention and treatment. Existing data analysis methods, however, still fail to give reliable and reproducible conclusions, because they insufficiently cope with the very specific data characteristics (high zero frequency, overdispersion, compositional, ...). We aim to (1) develop robust and flexible statistical methods for analysing clustered and longitudinal microbiome data, (2) adapt causal mediation analysis methods to work with microbiome data and integrate machine learning methods into the procedures, (3) develop informative visualisation tools for integrating microbiome data with other (high-dimensional) data sources, and (4) develop statistical joint models for integrating microbiome with other clinical outcome data for biomarker discovery. The project also involves a pilot study for studying the effect of the plant microbiome on the fecal and gut microbiome of mice and their effect on the immune response. Four PhD students will be involved and for each PhD project, at least two supervisors from at least two different institutes will be assigned.

Keywords:Immunology, Microbiology

Disciplines:Statistics, Data mining, Machine learning and decision making, Bio-informatics, High performance computing, Animal pathology, Data visualisation and high-throughput image analysis, Ecology not elsewhere classified, Microbiomes, Analysis of next-generation sequence data, Development of bioinformatics software, tools and databases, Immunology not elsewhere classified, Medical microbiomics, Microbiome