Gut microbiota ecology: Biodiversity estimated from hybrid neutral-niche model increases with health status and aging

differences or had the same statistical power. Our results suggest that estimating GM biodiver- sity through the ecological modeling that we proposed allows to derive biodiversity indices that are appropriate, i.e. consistent with other classical biodiversity indices, but enable to achieve higher statistical power and to unveil differences that would otherwise be masked by the intrinsic noise of the GM measurements. Materials and methods Data sets description, pre-processing and OTUs computation We considered 6 publicly available data sets containing 16S rRNA gene sequencing data of subjects for which the information about age and health status was available. We choose these data sets as they contained either a wide age range, data about aging-related diseases or data related to subjects with important diet and life-style differences. In the following, we describe the main features of each data set and the pre-processing pipeline that was used to obtain de novo Operational Taxonomic Units (OTUs). When processed sequencing data or OTUs were available, these were considered and eventually further processed. Notice that the pre-process- ing pipeline was adapted to each data set according to the choices performed by the original authors. Since the modelling and statistical analysis were performed separately for each data set, here the aim of the pre-processing is in fact to obtain clean data and not to obtain OTUs that are comparable between data sets. The main characteristics in terms of sequencing and

bioinformatic processing are summarized in S1 Table. The data sets characteristics in terms of age, sex, health status and numerosity of individuals included in the statistical analyses are summarized in S2 Table. i) ELDERMET: The ELDERMET data set [3, 34], includes 836 samples from 371 elderly (64–102 years old) and 13 young (26–46 years old) Irish subjects. Faecal samples were collected at 3 time points, approximately 3 months apart, that we will refer to as T0, T1 and T2. DNA was extracted from faecal samples, and sequence reads from 16S rRNA gene V4 amplicons were sequenced on a 454 Genome Sequencer FLX Titanium platform (Roche Diagnostics and Beckman Coulter Genomics) according to the manufacturer’s sequencing protocol. DNA sequences (fastq files) are available on the Sequence Read Archive under BioProject PRJNA283106. Besides age, available personal and clinical information for the elderly people include gender, antibiotics usage, Body Mass Index (BMI), calf circumference, residence set- ting, Mini Nutritional Assessment (MNA), Healthy Food Diversity index (HeFD), Functional Independence Measures (FIM), Mini Mental State Exam (MMSE), Barthel score, and values for interleukin (IL)-6, IL-8, IL-10 and Tumour Necrosis Factor (TNF)-α. Sequencing reads were filtered using fastx-toolkit [44] according to the following criteria: read length not shorter than 150 bp and not longer than 350 bp; no ambiguous bases (Ns); quality score higher than 25 in at least 50% of the read. OTUs were then obtained by clustering reads at 97% similarity using the UPARSE pipeline [45].

ii) Biagi & Schnorr: The data sets from Biagi et al. [33] and Schnorr et al. [35] include 17 Italian Down Syndrome (DS) persons and 16 age-matched Italian healthy young adults (20–40 years old) who adhered to the standard Mediterranean diet. The study from Schnorr et al. [35] also includes 27 Hadza hunter-gatherers (8–70 years old), whose diet is mainly based on meat, honey, baobab, berries and tubers. For all subjects, age is available but the match between the subject age and 16S rRNA data is not available. 16S rRNA gene V4 amplicons were sequenced on a 454 Genome Sequencer FLX Titanium platform (Roche Diagnostics and Beckman Coul- ter Genomics) according to the manufacturer’s sequencing protocol. DNA sequences (fasta files) are respectively available on MG-RAST under the project ids mgp10557 and mgp7058. Sequencing reads were filtered using mothur [46] according to the following criteria: read PLOS ONE Gut microbiota biodiversity and healthy aging PLOS ONE | https://doi.org/10.1371/journal.pone.0237207 October 30, 2020 13 / 23