Microbiome Preprocessing Machine Learning Pipeline.

Background: 16S sequencing results are often used for Machine Learning (ML) tasks. 16S gene sequences are represented as feature counts, which are associated with taxonomic representation. Raw feature counts may not be the optimal representation for ML. Methods: We checked multiple preprocessing steps and tested the optimal combination for 16S sequencing-based classification tasks. We computed the contribution of each step to the accuracy as measured by the Area Under Curve (AUC) of the classification. Results: We show that the log of the feature counts is much more informative than the relative counts. We further show that merging features associated with the same taxonomy at a given level, through a dimension reduction step for each group of bacteria improves the AUC. Finally, we show that z-scoring has a very limited effect on the results. Conclusions: The prepossessing of microbiome 16S data is crucial for optimal microbiome based Machine Learning. These preprocessing steps are integrated into the MIPMLP - Microbiome Preprocessing Machine Learning Pipeline, which is available as a stand-alone version at: https://github.com/louzounlab/microbiome/tree/master/Preprocess or as a service at http://mip-mlp.math.biu.ac.il/Home Both contain the code, and standard test sets.

Neonatal antibiotic exposure impairs child growth during the first six years of life by perturbing intestinal microbial colonization.

Exposure to antibiotics in the first days of life is thought to affect various physiological aspects of neonatal development. Here, we investigate the long-term impact of antibiotic treatment in the neonatal period and early childhood on child growth in an unselected birth cohort of 12,422 children born at full term. We find significant attenuation of weight and height gain during the first 6 years of life after neonatal antibiotic exposure in boys, but not in girls, after adjusting for potential confounders. In contrast, antibiotic use after the neonatal period but during the first 6 years of life is associated with significantly higher body mass index throughout the study period in both boys and girls. Neonatal antibiotic exposure is associated with significant differences in the gut microbiome, particularly in decreased abundance and diversity of fecal Bifidobacteria until 2 years of age. Finally, we demonstrate that fecal microbiota transplant from antibiotic-exposed children to germ-free male, but not female, mice results in significant growth impairment. Thus, we conclude that neonatal antibiotic exposure is associated with a long-term gut microbiome perturbation and may result in reduced growth in boys during the first six years of life while antibiotic use later in childhood is associated with increased body mass index.

Modulation of cytokine patterns and microbiome during pregnancy in IBD.

OBJECTIVE: Pregnancy may affect the disease course of IBD. Both pregnancy and IBD are associated with altered immunology and intestinal microbiology. However, to what extent immunological and microbial profiles are affected by pregnancy in patients with IBD remains unclear. DESIGN: Faecal and serum samples were collected from 46 IBD patients (31 Crohn's disease (CD) and 15 UC) and 179 healthy controls during first, second and third trimester of pregnancy, and prepregnancy and postpartum for patients with IBD. Peripheral blood cytokine profiles were determined by ELISA, and microbiome analysis was performed by sequencing the V4 region of the bacterial 16S rRNA gene. RESULTS: Proinflammatory serum cytokine levels in patients with IBD decrease significantly on conception. Reduced interleukin (IL)-10 and IL-5 levels but increased IL-8 and interferon (IFN)γ levels compared with healthy controls were seen throughout pregnancy, but cytokine patterns remained stable during gestation. Microbial diversity in pregnant patients with IBD was reduced compared with that in healthy women, and significant differences existed between patients with UC and CD in early pregnancy. However, these microbial differences were no longer present during middle and late pregnancy. Dynamic modelling showed considerable interaction between cytokine and microbial composition. CONCLUSION: Serum proinflammatory cytokine levels markedly improve on conception in pregnant patients with IBD, and intestinal microbiome diversity of patients with IBD normalises during middle and late pregnancy. We thus conclude that pregnancy is safe and even potentially beneficial for patients with IBD.

Progesterone Increases Bifidobacterium Relative Abundance during Late Pregnancy.

Gestation is accompanied by alterations in the microbial repertoire; however, the mechanisms driving these changes are unknown. Here, we demonstrate a dramatic shift in the gut microbial composition of women and mice during late pregnancy, including an increase in the relative abundance of Bifidobacterium. Using in-vivo-transplanted pellets, we found that progesterone, the principal gestation hormone, affects the microbial community. The effect of progesterone on the richness of several bacteria species, including Bifidobacterium, was also demonstrated in vitro, indicating a direct effect. Altogether, our results delineate a model in which progesterone promotes Bifidobacterium growth during late pregnancy.