Biodiversity in the era of big data. On the problem of taxonomy assignment and the distribution of diversity in complex biological systems

The study of complex biological matrices is a remarkable hot topic in biology. Soil, water, gut content are some of these matrices characterized by a prominent number of organisms living in tight connection. Hundreds or thousands of species and/or strains could be present in the same sample coming from different habitats (e.g. soil ecosystem) and showing inter-relationships, mainly energetic, to guarantee their ecosystem health functioning. The discrimination and/or identification of the different biological entities, at least for the eukaryotic components, using traditional morphological approaches is relatively complicated, requiring a specialist in each taxonomic groups and generally an appropriate long time to achieve a correct identification and classification. To overcome these limitations, molecular approaches have demonstrated to be valid alternatives where PCRs, cloning, DNA sequencing and bioinformatics analysis of sequence differences were used as the standard protocol. Nowadays, the genomic massive sequencing revolution, generated by the heterogeneous techniques collectively known as Next Generation Sequencing (NGS), has become the new gold standard. The present thesis consists of four sections which overpass detailed aspects of the analysis of biodiversity and the issues associated and elaborate both promises and pitfalls of coupling DNA barcode approach with high-throughput pyrosequencing in two different cases of biodiversity assessment. In the following a brief description of each section is provided: Section 1: Introduction to the biodiversity analysis problem. In this section an analysis of the main methods used to investigate the biodiversity and their related problems has been addressed. Emphasis has been put on the integrated approach between the classic DNA barcoding approach and the advantages of high processivity guaranteed by next generation sequencing technologies. Furthermore, the state of art regarding bioinformatics methods for species assignment and biodiversity patterns elaboration including phylogenetic diversity analysis are described. Section 2: Targeted Sequencing on Metazoan Communities. In this section, the precision and the accuracy of denoising procedure and the candidate parameters able to reduce sequence error rate are investigated. This work also proposed an innovative taxon assignment pipeline. In addition, a novel library preparation method allowing the sequencing of the entire coxI barcode region (approximately 700 bp) on 454 pyrosequencing platform (Roche Life Science) is proposed. To address the objectives, metazoan communities coming from complex environmental matrix (soil) were considered. Section 3: Microbiota invasion mediated by Varroa destructor to Apis mellifera. The starting hypothesis of this section is that varroa mites play a fundamental role in the alteration of bacterial composition of honey bee larvae, acting not only as a vector, but also as a sort of an open “door” through which exogenous bacteria alter the mechanisms of primary succession in the “simple” honey bee larval microbiome. To explore these dynamics a classical microbial communities analysis approach and a new approach considering the phylogenetic entropy as a measure of biodiversity were tested. The varroa and honey bee bacterial communities were studied through barcoded amplicon pyrosequencing method , taking advantage of the NGS methods and the opportunity to detect uncultured and uncultivable bacteria allowed by such techniques. Section 4: general conclusions and perspectives. General conclusions and future promises highlighted by the above mentioned experiments are illustrated in this section.