Cindy Boer

222 | Chapter 5.1 limitation when comparing different cohort datasets, is the accuracy of replication of 16S profiles, since sample collection, sample storage, DNA isolation, PCR amplification, amplified 16S rRNA variable region, the sequencing technique used and downstream bioinformatics may differ between different cohorts. For example, out of 33 associated OTUs with BMI, we could only replicate 6 in LLD and FGFP. In addition, geographical dif- ferences and lack of repetition may limit the accuracy of replicating microbiota profiles from different countries. In addition, although (besides using harmonized sample col- lection and dataset generation) we adjusted for the potential confounders in the anal- yses comparing children and adults, we should be aware that we cannot totally discard the presence of residual stratification as a consequence of the fact that both populations were derived from different cohorts. To conclude, in this study we performed microbiota profiling on the stools of 2,111 children in the age-range of 9 to 12 years and 1,427 adult individuals in the range of 46 to 88 years of age. We observed a clear distinction between the gut microbiomes of children as compared to adults, including differences in microbiota diversity and Fir- micutes and Bacteroidetes abundances. These changes were associated with predicted shifts in functional properties, including in energy metabolism, antibiotic production and the production of essential B-vitamins. These observations are likely due to the development of human-gut microbiome interactions with age. As both GenR and the RS cohort have been deeply characterized[24,25]. We here presented two valuable data- sets for studying the possible association of the human stool microbiome and, life style, environmental factors, and health and disease outcomes. In addition, as the participants of both cohorts have been genotypes, also association between host genetics and host microbiome could be investigated. Acknowledgements The generation and management of stool microbiome data for the Generation R Study (Focus @9) and the Rotterdam Study (RSIII-2) was executed by the Human Genotyp- ing Facility of the Genetic Laboratory of the Department of Internal Medicine, Erasmus MC, Rotterdam, The Netherlands. We thank Nahid El Faquir and Jolande Verkroost-Van Heemst for their help in sample collection and registration, and Kamal Arabe, Hedayat Razawy and Karan Singh Asra for their help in DNA isolation and sequencing. Further- more, we thank drs. Jeroen Raes and Jun Wang (KU Leuven, Belgium) for their guidance in 16S rRNA profiling and dataset generation. Djawad Radjabzadeh was funded by an Erasmus MC mRACE grant “Profiling of the human gut microbiome”. The authors are very grateful to the study participants of the Rotterdam Study, the staff from the Rotter-