Il ricino (Ricinus communis L.) è una tra le dieci principali colture oleaginose a livello mondiale. La ricina, eliminata nei processi industriali per la produzione di olio di ricino, costituisce un rischio sia per lo sfruttamento della pianta come biomassa per la generazione di biocarburanti vegetali di seconda generazione, sia per la possibilità di ottenere scorie tossiche dalla lavorazione del materiale. L’utilizzo di tecniche di mutagenesi chimica, associata ad AFLP e metodiche di sequenziamento bersaglio specifiche (TILLING) hanno dimostrato la possibilità di ottenere e riconoscere, in tempi relativamente brevi e in modo specifico, mutazioni nel genoma d’interesse. A partire da una popolazione monovarietale di ricino, sono stati messi a punto diversi trattamenti (basati su EMS ed MNU), al fine di ottenere una popolazione mutagenizzata chimicamente al cui interno ricercare piante prive, o con un contenuto ridotto, di ricina nel seme e di conseguenza caratterizzate da un maggior valore agronomico ed economico. In seguito all’autoimpollinazione delle piante sopravvissute, analisi AFLP e di sequenziamento sulle generazioni successive hanno mostrato come le piante tendano ad accumulare variazioni nel loro genoma rispetto alla generazione precedente. Dall’analisi di sequenza di circa 1 Mb del gene per la ricina nella popolazione mutagenizzata non è emerso nessun cambiamento nella sequenza nucleotidica, in accordo con i risultati di altri studi in cui si è visto come la frequenza di mutagenesi possa variare con la specie considerata. / Castor (Ricinus communis L.) is one of the ten major oil crops worldwide. Ricin, eliminated in industrial processes for the production of castor oil, constitutes a risk to both the exploitation of the plant as a biomass to generate second-generation biofuels, both for getting toxic waste from the processing of the material. The use of chemical mutagenesis techniques, AFLP and methods associated with specific target sequences (TILLING) have demonstrated the ability to obtain and recognize, in a relatively short time, and specifically, mutations in the genome of interest. Starting from a population of castor-variety, have been developed different treatments (based on EMS and MNU), in order to obtain a chemically mutagenized population of plants with a reduced content of ricin in the seed and therefore characterized by a greater agronomic and economic value. After self-pollination of the survived plants, AFLP analysis and sequencing showed how plants tend to accumulate changes in their genome than the previous generation. A sequence analysis of about 1 Mb of the gene for ricin in mutagenized populations has revealed no change in the nucleotide sequence, in agreement with results of other studies in which we have seen that the frequency of mutation may vary with the species under consideration.

The goal of this thesis is the discovery of a bioinformatics solution for network-based predictive analysis of NGS data, in which network structures can substitute gene lists as a more rich and complex signature of disease. I have focused on methods for network stability, network inference and network comparison, as additional components of the pipeline and as methods to detects outliers in high-throughput datasets. Besides a first work on GEO datasets, the main application of my pipeline has been on original data from the FDA SEQC (Sequencing Quality Control)project. Here I will report some initial findings to which I have contributed with methods and analysis: as the corresponding papers are being submitted. My goal is to provide a comprehensive tool for network reconstruction and network comparison as an R package and user-friendly web service interface available on-line at https://renette.fbk.eu The goal of this thesis is the discovery of a bioinformatics solution for network-based predictive analysis of NGS data, in which network structures can substitute gene lists as a more rich and complex signature of disease. I have focused on methods for network stability, network inference and network comparison, as additional components of the pipeline and as methods to detects outliers in high-throughput datasets. Besides a first work on GEO datasets, the main application of my pipeline has been on original data from the FDA SEQC (Sequencing Quality Control)project. Here I will report some initial findings to which I have contributed with methods and analysis: as the corresponding papers are being submitted. My goal is to provide a comprehensive tool for network reconstruction and network comparison as an R package and user-friendly web service interface available on-line at https://renette.fbk.eu.

In recent years, the world has faced major health challenges, from the rise of antibiotic resistance to the emergence of new pathogens with pandemic potential. This highlights the importance of considering human health as inextricably intertwined with that of other animals and the environment in which they live. This paradigm is known as “One Health”, which is the integration of environmental sciences, veterinary science, and medicine. Within this perspective, viruses, the most abundant biological entities on Earth, play a central role in connecting different organisms, deeply influencing the health of their hosts. Despite their great importance, most viruses are still poorly understood, mainly because of the technical and economic limitations posed by isolation, cultivation, and single colony sequencing. However, recently developed genomic technologies offer a cheaper and more sensitive alternative to study viruses, allowing a better integration of data from various sources and making it possible to explore how they circulate among different hosts and environments. In this thesis, I hypotesise that, by combining different classes of genomic methods with One Health practices, it is possible to reveal much more of the entire picture of viral diversity and evolution that by simply using them in a separated way. I show that this is the case for each one of the scientific questions addressed in this work, which are organised in three main chapters: - In the second chapter I analysed 22 metatranscriptomes from tick samples from different parts of Italy, obtaining a set of 91 viral contigs for which I reconstructed the phylogeny, with the aim to identify the presence of possible pathogens and characterise the unexplored viral diversity in the country. This analysis not only clarified the molecular epidemiology of well-known pathogens such as Tick-borne encephalitis virus, but also allowed the discovery of at least 10 novel viral species. - In the third chapter, I investigated the origin and spread of West Nile virus, an emerging pathogen causing neurological disease worldwide. The goal was to expand the current knowledge of this virus by increasing the number of sequenced genomes and to reconstruct how the virus moved between Africa and Europe. Phylogeographic and phylodynamic methods showed that this pathogen originated in Africa and repeatedly invaded the European continent, revealing the dynamics of its evolution through space and time. - In the last chapter, I contributed to obtaining a set of DNA phages assembled from human microbiomes and viromes by manual curation and comparative genomics and developed a new approach to study their evolution in relation to their bacterial hosts. I found that the strength of co-phylogenetic patterns between viruses and their hosts is generally weak, suggesting that their ecological relationships emerge on short evolutionary timescales. Taken together, these results show how the integration of viral metagenomics and phylogenomics in One Health is essential to answer fundamental questions about the diversity of viruses and how they spread and evolve. Furthermore, the methods and protocols developed in these studies can be applied to similar cases, allowing a systematic exploration of many other datasets to expand our knowledge of the virosphere. This information can be used to implement containment strategies, public health policies, therapies, and biotechnologies.