We developed highly innovative computational methods to study microbial communities at unprecedented depth and resolution, based on the growing availability of microbiome data being scalable to many thousands of samples and thus considered "big data". The computational tools were published in scientific journals, presented at scientific conferences, taught in classes and courses, and are open source and publicly available with guides and tutorials.
With the new tools developed, we analyzed thousands of microbiome samples from multiple existing studies that have been curated and integrated within MetaPG. This generated massive information about the population genomics of hundreds of species that were previously completely or almost completely neglected. We investigated association between specific strains/subspecies with conditions of interest and produced several large comparative meta-analysis, processing more than 10,000 samples and over 150,000 reconstructed microbial genomes. Among the investigations, we analysed metagenomes from various countries, body sites, ages, and lifestyles, identifying thousands of microbial genomes from yet-to-be-named species and expanding the knowledge of the human-associated microbes. In a meta-analysis of the gut microbiome in colorectal cancer patients, we identified reproducible microbiome biomarkers and accurate disease-predictive models. We performed a cross-continent meta-analysis exploiting >6,500 metagenomes.
By the large scale comparative meta-analysis, we identified and characterized microbes that are undergoing extinction in the Westernized world using newly sequenced non-Westerninzed cohorts and publicly available data. Three main findings: i) Prevotella copri, previously considered just one microbial species, is a complex composed of four distinct groups and was found more prevalent in non-Westernized populations probably due to diet\lifestyle; ii) thousands of new human microbiome species were uncovered by a large-scale analysis, mostly associated to non-Westernized populations; iii) the study on Blastocystis, a commonly colonizer of the gut, detected different subtypes and extended previous observations on its high prevalence in the population of non-Westernized individuals; iv) an investigation on Eubacterium rectale, one of the most prevalent human gut bacteria, identified possible association of its distribution and human migration along time.
By the end of the project, we managed to achieve all the project objectives and to provide outcomes that will lead to new crucial insights into host-microbe and microbe-microbe interaction dynamics and, potentially, also to consider these strains as probiotic products in the future. Within MetaPG, we published 43 peer-reviewed papers in high standard scientific Journals, with over 1500 citations so far.
