Extra steps by Member States are needed to protect farmed and grassland ecosystems

To systematically develop dietary strategies based on resistant starch (RS) that modulate the human gut microbiome, detailed in vivo studies that evaluate the effects of different forms of RS on the community structure and population dynamics of the gut microbiota are necessary. The aim of the present study was to gain a community wide perspective of the effects of RS types 2 (RS2) and 4 (RS4) on the fecal microbiota in human individuals. Ten human subjects consumed crackers for three weeks each containing either RS2, RS4, or native starch in a double-blind, crossover design. Multiplex sequencing of 16S rRNA tags revealed that both types of RS induced several significant compositional alterations in the fecal microbial populations, with differential effects on community structure. RS4 but not RS2 induced phylum-level changes, significantly increasing Actinobacteria and Bacteroidetes while decreasing Firmicutes. At the species level, the changes evoked by RS4 were increases in Bifidobacterium adolescentis and Parabacteroides distasonis, while RS2 significantly raised the proportions of Ruminococcus bromii and Eubacterium rectale when compared to RS4. The population shifts caused by RS4 were numerically substantial for several taxa, leading for example, to a ten-fold increase in bifidobacteria in three of the subjects, enriching them to 18–30% of the fecal microbial community. The responses to RS and their magnitudes varied between individuals, and they were reversible and tightly associated with the consumption of RS. Our results demonstrate that RS2 and RS4 show functional differences in their effect on human fecal microbiota composition, indicating that the chemical structure of RS determines its accessibility by groups of colonic bacteria. The findings imply that specific bacterial populations could be selectively targeted by well designed functional carbohydrates, but the inter-subject variations in the response to RS indicates that such strategies might benefit from more personalized approaches.

Balancing model complexity is a key challenge of modern computational ecology, particularly so since the spread of machine learning algorithms. Species distribution models are often implemented using a wide variety of machine learning algorithms that can be fine‐tuned to achieve the best model prediction while avoiding overfitting. We have released SDMtune, a new R package that aims to facilitate training, tuning, and evaluation of species distribution models in a unified framework. The main innovations of this package are its functions to perform data‐driven variable selection, and a novel genetic algorithm to tune model hyperparameters. Real‐time and interactive charts are displayed during the execution of several functions to help users understand the effect of removing a variable or varying model hyperparameters on model performance. SDMtune supports three different metrics to evaluate model performance: the area under the receiver operating characteristic curve, the true skill statistic, and Akaike's information criterion corrected for small sample sizes. It implements four statistical methods: artificial neural networks, boosted regression trees, maximum entropy modeling, and random forest. Moreover, it includes functions to display the outputs and create a final report. SDMtune therefore represents a new, unified and user‐friendly framework for the still‐growing field of species distribution modeling.