Roteiro Tecnológico (roadmap) da Captação, Utilização e Armazenamento de Dióxido de Carbono (CCUS) em Portugal

O objetivo primordial deste trabalho foi estabelecer um roteiro tecnológico para aplicação das tecnologias de “Captação, Utilização e Sequestração de Carbono - CCUS” em Portugal. Para o efeito procedeu-se à identificação da origem das maiores fontes emissoras estacionárias industriais de CO2, adotando como critério o valor mínimo de 1×105 ton CO2/ano e limitado apenas ao território continental. Com base na informação recolhida e referente aos dados oficiais mais recentes (ano de 2013), estimou-se que o volume de emissões industriais de CO2 possível de captar em Portugal, corresponde a cerca de 47 % do valor global das emissões industriais, sendo oriundo de três setores de atividade industrial: produção de cimento, de pasta de papel e centrais termoelétricas a carvão. A maioria das grandes fontes emissoras industriais localiza-se no litoral do país, concentrando-se entre Aveiro e Sines. Pelas condicionantes geográficas do país e, sobretudo pela vantagem de já existir uma rede de gasodutos para o transporte de gás natural, com as respetivas infraestruturas de apoio associadas, admitiu-se que o cenário mais favorável para o transporte do CO2 captado será a criação de um sistema de transporte por gasoduto específico para o CO2. Como critério de compatibilização da proximidade das fontes emissoras de CO2 com potenciais locais para o armazenamento geológico das correntes captadas, adotou-se a distância máxima de 100 km, considerada adequada perante a dimensão do território nacional e as características do tecido industrial nacional. Efetuou-se a revisão das tecnologias de captação de CO2 disponíveis, quer comercialmente, quer em níveis avançados de demonstração e procedeu-se à análise exploratória da adequação desses diferentes métodos de captação a cada um dos setores de atividade industrial previamente identificados com emissões de CO2 suscetíveis de serem captadas. Na perspetiva da melhor integração dos processos, esta análise preliminar tomou em consideração as características das misturas gasosas, assim como o contexto industrial correspondente e o processo produtivo que lhe dá origem. As possibilidades de utilização industrial do CO2 sujeito à captação no país foram tratadas neste trabalho de forma genérica dado que a identificação de oportunidades reais para a utilização de correntes de CO2 captadas exige uma análise de compatibilização das necessidades efetivas de utilização de CO2 por parte de potenciais utilizadores industriais que carece da caracterização prévia das propriedades dessas correntes. Este é um tipo de análise muito específico que pressupõe o interesse mútuo de diferentes intervenientes: agentes emissores de CO2, operadores de transporte e, principalmente, potenciais utilizadores de CO2 como: matéria-prima para a síntese de compostos, solvente de extração supercrítica na indústria alimentar ou farmacêutica, agente corretor de pH em tratamento de efluentes, biofixação por fotossíntese, ou outra das aplicações possíveis identificadas para o CO2 captado. A última etapa deste estudo consistiu na avaliação das possibilidades de armazenamento geológico do CO2 captado e envolveu a identificação, nas bacias sedimentares nacionais, de formações geológicas com características reconhecidas como sendo boas indicações para o armazenamento de CO2 de forma permanente e em segurança. Seguiu-se a metodologia preconizada por organizações internacionais aplicando à situação nacional, critérios de seleção e de segurança que se encontram reconhecidamente definidos. A adequação para o armazenamento de CO2 das formações geológicas pré-selecionadas terá que ser comprovada por estudos adicionais que complementem os dados já existentes sobre as características geológicas destas formações e, mais importante ainda, por testes laboratoriais e ensaios de injeção de CO2 que possam fornecer informação concreta para estimar a capacidade de sequestração e de retenção de CO2 nestas formações e estabelecer os modelos geológicos armazenamento que permitam identificar e estimar, de forma concreta e objetiva, os riscos associados à injeção e armazenamento de CO2.

Identification and characterization of parasitism genes from the pinewood nematode Bursaphelenchus xylophilus reveals a multilayered detoxification strategy.

The migratory endoparasitic nematode Bursaphelenchus xylophilus, which is the causal agent of pine wilt disease, has phytophagous and mycetophagous phases during its life cycle. This highly unusual feature distinguishes it from other plantparasitic nematodes and requires profound changes in biology between modes. During the phytophagous stage, the nematode migrates within pine trees, feeding on the contents of parenchymal cells. Like other plant pathogens, B. xylophilus secretes effectors from pharyngeal gland cells into the host during infection.We provide the first description of changes in the morphology of these gland cells between juvenile and adult life stages. Using a comparative transcriptomics approach and an effector identification pipeline, we identify numerous novel parasitism genes which may be important for the mediation of interactions of B. xylophilus with its host. In-depth characterization of all parasitism genes using in situ hybridization reveals two major categories of detoxification proteins, those specifically expressed in either the pharyngeal gland cells or the digestive system. These data suggest that B. xylophilus incorporates effectors in a multilayer detoxification strategy in order to protect itself from host defence responses during phytophagy.

PROPUESTA METODOLÓGICA PARA LA ZONIFICACIÓN DE LA VULNERABILIDAD ESTRUCTURAL Y DE LA POBLACIÓN ANTE LA PRESENCIA DE UNA AMENAZA TECNOLÓGICA:EL CASO DEL POLIDUCTO DE RECOPE, COSTA RICA

Este trabajo crea y aplica una metodología acorde con la realidad del país para zonificar la vulnerabilidad estructural y de la población sometida a una amenaza tecnológica, en este caso bajo la influencia de una posible emergencia por derrame de combustibles del poliducto.  Una de las finalidades del trabajo es zonificar esta vulnerabilidad como elemento por considerar en la determinación del riesgo, elaboración de planes municipales de contingencia y en las propuestas de ordenamiento territorial.Palabras claves: Poliducto, RECOPE, vulnerabilidad, riesgo, derrame de combustibles, amenaza tecnológica.Abstract: This study develops and applies a methodology in keeping with the reality of the country in order to zone the structural and population vulnerability subject to technological hazard, in this case the influence of a possible emergency due to fuel spill from an oil pipeline. One of the objectives of this study is to zone this vulnerability as an element to consider in determining risk, devising municipal contingency plans and in proposing territorial coding.Keywords: Pipeline, RECOPE, vulnerability, risk, fuel spill, hazard, technological hazard.

Development of Mass Spectrometry-based analytical tools for characterization of bioconjugate vaccines in E. coli

Antigen design is generally driven by the need to obtain enhanced stability,efficiency and safety in vaccines.Unfortunately,the antigen modification is rarely proceeded in parallel with analytical tools development characterization.The analytical tools set up is required during steps of vaccine manufacturing pipeline,for vaccine production modifications,improvements or regulatory requirements.Despite the relevance of bioconjugate vaccines,robust and consistent analytical tools to evaluate the extent of carrier glycosylation are missing.Bioconjugation is a glycoengineering technology aimed to produce N-glycoprotein in vivo in E.coli cells,based on the PglB-dependent system by C. jejuni,applied for production of several glycoconjugate vaccines.This applicability is due to glycocompetent E. coli ability to produce site-selective glycosylated protein used,after few purification steps, as vaccines able to elicit both humoral and cell-mediate immune-response.Here, S.aureus Hla bioconjugated with CP5 was used to perform rational analytical-driven design of the glycosylation sites for the glycosylation extent quantification by Mass Spectrometry.The aim of the study was to develop a MS-based approach to quantify the glycosylation extent for in-process monitoring of bioconjugate production and for final product characterization.The three designed consensus sequences differ for a single amino-acid residue and fulfill the prerequisites for engineered bioconjugate more appropriate from an analytical perspective.We aimed to achieve an optimal MS detectability of the peptide carrying the consensus sequences,complying with the well-characterized requirements for N-glycosylation by PglB.Hla carrier isoforms,bearing these consensus sequences allowed a recovery of about 20 ng/μg of periplasmic protein glycosylated at 40%.The SRM-MS here developed was successfully applied to evaluate the differential site occupancy when carrier protein present two glycosites.The glycosylation extent in each glycosite was determined and the difference in the isoforms were influenced either by the overall source of protein produced and by the position of glycosite insertion.The analytical driven design of the bioconjugated antigen and the development of accurate,precise and robust analytical method allowed to finely characterize the vaccine.

Design and Development of Network Monitoring Strategies in P4-enabled Programmable Switches

Network monitoring is of paramount importance for effective network management: it allows to constantly observe the network’s behavior to ensure it is working as intended and can trigger both automated and manual remediation procedures in case of failures and anomalies. The concept of SDN decouples the control logic from legacy network infrastructure to perform centralized control on multiple switches in the network, and in this context, the responsibility of switches is only to forward packets according to the flow control instructions provided by controller. However, as current SDN switches only expose simple per-port and per-flow counters, the controller has to do almost all the processing to determine the network state, which causes significant communication overhead and excessive latency for monitoring purposes. The absence of programmability in the data plane of SDN prompted the advent of programmable switches, which allow developers to customize the data-plane pipeline and implement novel programs operating directly in the switches. This means that we can offload certain monitoring tasks to programmable data planes, to perform fine-grained monitoring even at very high packet processing speeds. Given the central importance of network monitoring exploiting programmable data planes, the goal of this thesis is to enable a wide range of monitoring tasks in programmable switches, with a specific focus on the ones equipped with programmable ASICs. Indeed, most network monitoring solutions available in literature do not take computational and memory constraints of programmable switches into due account, preventing, de facto, their successful implementation in commodity switches. This claims that network monitoring tasks can be executed in programmable switches. Our evaluations show that the contributions in this thesis could be used by network administrators as well as network security engineers, to better understand the network status depending on different monitoring metrics, and thus prevent network infrastructure and service outages.

Robotic Sensing and Manipulation of Deformable Linear Objects with Learning-based methods

Nowadays robotic applications are widespread and most of the manipulation tasks are efficiently solved. However, Deformable-Objects (DOs) still represent a huge limitation for robots. The main difficulty in DOs manipulation is dealing with the shape and dynamics uncertainties, which prevents the use of model-based approaches (since they are excessively computationally complex) and makes sensory data difficult to interpret. This thesis reports the research activities aimed to address some applications in robotic manipulation and sensing of Deformable-Linear-Objects (DLOs), with particular focus to electric wires. In all the works, a significant effort was made in the study of an effective strategy for analyzing sensory signals with various machine learning algorithms. In the former part of the document, the main focus concerns the wire terminals, i.e. detection, grasping, and insertion. First, a pipeline that integrates vision and tactile sensing is developed, then further improvements are proposed for each module. A novel procedure is proposed to gather and label massive amounts of training images for object detection with minimal human intervention. Together with this strategy, we extend a generic object detector based on Convolutional-Neural-Networks for orientation prediction. The insertion task is also extended by developing a closed-loop control capable to guide the insertion of a longer and curved segment of wire through a hole, where the contact forces are estimated by means of a Recurrent-Neural-Network. In the latter part of the thesis, the interest shifts to the DLO shape. Robotic reshaping of a DLO is addressed by means of a sequence of pick-and-place primitives, while a decision making process driven by visual data learns the optimal grasping locations exploiting Deep Q-learning and finds the best releasing point. The success of the solution leverages on a reliable interpretation of the DLO shape. For this reason, further developments are made on the visual segmentation.

Development of a framework for the analysis of document-based phishing campaigns

In questo elaborato di tesi si affronta lo sviluppo di un framework per l'analisi di URL di phishing estratte da documenti malevoli. Tramite il linguaggio python3 e browsers automatizzati si è sviluppata una pipeline per analizzare queste campagne malevole. La pipeline ha lo scopo di arrivare alla pagina finale, evitando di essere bloccata da tecniche anti-bot di cloaking, per catturare una schermata e salvare la pagina in locale. Durante l'analisi tutto il traffico è salvato per analisi future. Ad ogni URL visitato vengono salvate informazioni quali entry DNS, codice di Autonomous System e lo stato nella blocklist di Google. Un'analisi iniziale delle due campagne più estese è stata effettuata, rivelando il business model dietro ad esse e le tecniche usate per proteggere l'infrastruttura stessa.

Supporting Scientific Research Through Machine and Deep Learning: Fluorescence Microscopy and Operational Intelligence Use Cases

Although the debate of what data science is has a long history and has not reached a complete consensus yet, Data Science can be summarized as the process of learning from data. Guided by the above vision, this thesis presents two independent data science projects developed in the scope of multidisciplinary applied research. The first part analyzes fluorescence microscopy images typically produced in life science experiments, where the objective is to count how many marked neuronal cells are present in each image. Aiming to automate the task for supporting research in the area, we propose a neural network architecture tuned specifically for this use case, cell ResUnet (c-ResUnet), and discuss the impact of alternative training strategies in overcoming particular challenges of our data. The approach provides good results in terms of both detection and counting, showing performance comparable to the interpretation of human operators. As a meaningful addition, we release the pre-trained model and the Fluorescent Neuronal Cells dataset collecting pixel-level annotations of where neuronal cells are located. In this way, we hope to help future research in the area and foster innovative methodologies for tackling similar problems. The second part deals with the problem of distributed data management in the context of LHC experiments, with a focus on supporting ATLAS operations concerning data transfer failures. In particular, we analyze error messages produced by failed transfers and propose a Machine Learning pipeline that leverages the word2vec language model and K-means clustering. This provides groups of similar errors that are presented to human operators as suggestions of potential issues to investigate. The approach is demonstrated on one full day of data, showing promising ability in understanding the message content and providing meaningful groupings, in line with previously reported incidents by human operators.

Whole Exome Sequencing widens the genetic landscape of Hereditary Optic Neuropathies

Hereditary optic neuropathies (HON) are a genetic cause of visual impairment characterized by degeneration of retinal ganglion cells. The majority of HON are caused by pathogenic variants in mtDNA genes and in gene OPA1. However, several other genes can cause optic atrophy and can only be identified by high throughput genetic analysis. Whole Exome Sequencing (WES) is becoming the primary choice in rare disease molecular diagnosis, being both cost effective and informative. We performed WES on a cohort of 106 cases, of which 74 isolated ON patients (ON) and 32 syndromic ON patients (sON). The total diagnostic yield amounts to 27%, slightly higher for syndromic ON (31%) than for isolated ON (26%). The majority of genes found are related to mitochondrial function and already reported for harbouring HON pathogenic variants: ACO2, AFG3L2, C19orf12, DNAJC30, FDXR, MECR, MTFMT, NDUFAF2, NDUFB11, NDUFV2, OPA1, PDSS1, SDHA, SSBP1, and WFS1. Among these OPA1, ACO2, and WFS1 were confirmed as the most relevant genetic causes of ON. Moreover, several genes were identified, especially in sON patients, with direct impairment of non-mitochondrial molecular pathways: from autophagy and ubiquitin system (LYST, SNF8, WDR45, UCHL1), to neural cells development and function (KIF1A, GFAP, EPHB2, CACNA1A, CACNA1F), but also vitamin metabolism (SLC52A2, BTD), cilia structure (USH2A), and nuclear pore shuttling (NUTF2). Functional validation on yeast model was performed for pathogenic variants detected in MECR, MTFMT, SDHA, and UCHL1 genes. For SDHA and UCHL1 also muscle biopsy and fibroblast cell lines from patients were analysed, pointing to possible pathogenic mechanisms that will be investigated in further studies. In conclusion, WES proved to be an efficient tool when applied to our ON cohort, for both common disease-genes identification and novel genes discovery. It is therefore recommended to consider WES in ON molecular diagnostic pipeline, as for other rare genetic diseases.

Contribution of rare damaging variants in Autism Spectrum Disorder

Autism Spectrum Disorder (ASD) is a heterogeneous and highly heritable neurodevelopmental disorder with a complex genetic architecture, consisting of a combination of common low-risk and more penetrant rare variants. This PhD project aimed to explore the contribution of rare variants in ASD susceptibility through NGS approaches in a cohort of 106 ASD families including 125 ASD individuals. Firstly, I explored the contribution of inherited rare variants towards the ASD phenotype in a girl with a maternally inherited pathogenic NRXN1 deletion. Whole exome sequencing of the trio family identified an increased burden of deleterious variants in the proband that could modulate the CNV penetrance and determine the disease development. In the second part of the project, I investigated the role of rare variants emerging from whole genome sequencing in ASD aetiology. To properly manage and analyse sequencing data, a robust and efficient variant filtering and prioritization pipeline was developed, and by its application a stringent set of rare recessive-acting and ultra-rare variants was obtained. As a first follow-up, I performed a preliminary analysis on de novo variants, identifying the most likely deleterious variants and highlighting candidate genes for further analyses. In the third part of the project, considering the well-established involvement of calcium signalling in the molecular bases of ASD, I investigated the role of rare variants in voltage-gated calcium channels genes, that mainly regulate intracellular calcium concentration, and whose alterations have been correlated with enhanced ASD risk. Specifically, I functionally tested the effect of rare damaging variants identified in CACNA1H, showing that CACNA1H variation may be involved in ASD development by additively combining with other high risk variants. This project highlights the challenges in the analysis and interpretation of variants from NGS analysis in ASD, and underlines the importance of a comprehensive assessment of the genomic landscape of ASD individuals.

High performance and energy-efficient instruction cache design and optimisation for ultra-low-power multi-core clusters

High Energy efficiency and high performance are the key regiments for Internet of Things (IoT) end-nodes. Exploiting cluster of multiple programmable processors has recently emerged as a suitable solution to address this challenge. However, one of the main bottlenecks for multi-core architectures is the instruction cache. While private caches fall into data replication and wasting area, fully shared caches lack scalability and form a bottleneck for the operating frequency. Hence we propose a hybrid solution where a larger shared cache (L1.5) is shared by multiple cores connected through a low-latency interconnect to small private caches (L1). However, it is still limited by large capacity miss with a small L1. Thus, we propose a sequential prefetch from L1 to L1.5 to improve the performance with little area overhead. Moreover, to cut the critical path for better timing, we optimized the core instruction fetch stage with non-blocking transfer by adopting a 4 x 32-bit ring buffer FIFO and adding a pipeline for the conditional branch. We present a detailed comparison of different instruction cache architectures' performance and energy efficiency recently proposed for Parallel Ultra-Low-Power clusters. On average, when executing a set of real-life IoT applications, our two-level cache improves the performance by up to 20% and loses 7% energy efficiency with respect to the private cache. Compared to a shared cache system, it improves performance by up to 17% and keeps the same energy efficiency. In the end, up to 20% timing (maximum frequency) improvement and software control enable the two-level instruction cache with prefetch adapt to various battery-powered usage cases to balance high performance and energy efficiency.

The environmental DNA in the risk assessment and decision making processes for the invasive species management in agri-food sector, hydraulic security and biodiversity conservation

The use of environmental DNA (eDNA) analysis as a monitoring tool is becoming more and more widespread. The eDNA metabarcoding methods allow rapid community assessments of different target taxa. This work is focused on the validation of the environmental DNA metabarcoding protocol for biodiversity assessment of freshwater habitats. Scolo Dosolo was chosen as study area and three sampling points were defined for traditional and eDNA analyses. The gutter is a 205 m long anthropic canal located in Sala Bolognese (Bologna, Italy). Fish community and freshwater invertebrate metazoans were the target groups for the analysis. After a preliminary study in summer 2019, 2020 was devoted to the sampling campaign with winter (January), spring (May), summer (July) and autumn (October) surveys. Alongside with the water samplings for the eDNA study, also traditional fish surveys using the electrofishing technique were performed to assess fish community composition; census on invertebrates was performed using an entomological net and a surber sampler. After in silico analysis, the MiFish primer set amplifying a fragment of the 12s rRNA gene was selected for bony fishes. For invertebrates the FWHF2 + FWHR2N primer combination, that amplifies a region of the mitochondrial coi gene, was chosen. Raw reads were analyzed through a bioinformatic pipeline based on OBITools metabarcoding programs package and QIIME2. The OBITools pipeline retrieved seven fish taxa and 54 invertebrate taxa belonging to six different phyla, while QIIME2 recovered eight fish taxa and 45 invertebrate taxa belonging to the same six phyla as the OBITools pipeline. The metabarcoding results were then compared with the traditional surveys data and bibliographic records. Overall, the validated protocol provides a reliable picture of the biodiversity of the study area and an efficient support to the traditional methods.

Structural and biophysical characterization of RAD52 as novel pharmacological target for synthetic lethality

RAD52 is a protein involved in various DNA reparation mechanisms. In the last few years, RAD52 has been proposed as a novel pharmacological target for cancer synthetic lethality strategies. Hence, this work has the purpose to investigate RAD52 protein, with biophysical and structural tools to shed light on proteins features and mechanistic details that are, up to now poorly described, and to design novel strategies for its inhibition. My PhD work had two goals: the structural and functional characterization of RAD52 and the identification of novel RAD52 inhibitors. For the first part, RAD52 was characterized both for its DNA interaction and oligomerization state together with its propensity to form high molecular weight superstructures. Moreover, using EM and Cryo-EM techniques, additional RAD52 structural hallmarks were obtained, valuable both for understanding protein mechanism of action and for drug discovery purpose. The second part of my PhD project focused on the design and characterization of novel RAD52 inhibitors to be potentially used in combination therapies with PARPi to achieve cancer cells synthetic lethality, avoiding resistance occurrence and side effects. With this aim we selected and characterized promising RAD52 inhibitors through three different approaches: 19F NMR fragment-based screening; virtual screening campaign; aptamers computational design. Selected hits (fragments, molecules and aptamers) were investigated for their binding to RAD52 and for their mechanism of inhibition. Collected data highlighted the identification of hits worthy to be developed into more potent and selective RAD52 inhibitors. Finally, a side project carried out during my PhD is reported. GSK-3β protein, an already validated pharmacological target was investigated using biophysical and structural biology tools. Here, an innovative and adaptable drug discovery screening pipeline able to directly identify selective compounds with binding affinities not higher than a reference binder was developed.

Towards a full cosmological exploitation of cosmic voids

A stately fraction of the Universe volume is dominated by almost empty space. Alongside the luminous filamentary structures that make it up, there are vast and smooth regions that have remained outside the Cosmology spotlight during the past decades: cosmic voids. Although essentially devoid of matter, voids enclose fundamental information about the cosmological framework and have gradually become an effective and competitive cosmological probe. In this Thesis work we present fundamental results about the cosmological exploitation of voids. We focused on the number density of voids as a function of their radius, known as void size function, developing an effective pipeline for its cosmological usage. We proposed a new parametrisation of the most used theoretical void size function to model voids identified in the distribution of biased tracers (i.e. dark matter haloes, galaxies and galaxy clusters), a step of fundamental importance to extend the analysis to real data surveys. We then applied our built methodology to study voids in alternative cosmological scenarios. Firstly we exploited voids with the aim of breaking the degeneracies between cosmological scenarios characterised by modified gravity and the inclusion of massive neutrinos. Secondly we analysed voids in the perspective of the Euclid survey, focusing on the void abundance constraining power on dynamical dark energy models with massive neutrinos. Moreover we explored other void statistics like void profiles and clustering (i.e. the void-galaxy and the void-void correlation), providing cosmological forecasts for the Euclid mission. We finally focused on the probe combination, highlighting the incredible potential of the joint analysis of multiple void statistics and of the combination of the void size function with different cosmological probes. Our results show the fundamental role of the void analysis in constraining the fundamental parameters of the cosmological model and pave the way for future studies on this topic.

Deploying and processing neural representations of signals

Neural representations (NR) have emerged in the last few years as a powerful tool to represent signals from several domains, such as images, 3D shapes, or audio. Indeed, deep neural networks have been shown capable of approximating continuous functions that describe a given signal with theoretical infinite resolution. This finding allows obtaining representations whose memory footprint is fixed and decoupled from the resolution at which the underlying signal can be sampled, something that is not possible with traditional discrete representations, e.g., grids of pixels for images or voxels for 3D shapes. During the last two years, many techniques have been proposed to improve the capability of NR to approximate high-frequency details and to make the optimization procedures required to obtain NR less demanding both in terms of time and data requirements, motivating many researchers to deploy NR as the main form of data representation for complex pipelines. Following this line of research, we first show that NR can approximate precisely Unsigned Distance Functions, providing an effective way to represent garments that feature open 3D surfaces and unknown topology. Then, we present a pipeline to obtain in a few minutes a compact Neural Twin® for a given object, by exploiting the recent advances in modeling neural radiance fields. Furthermore, we move a step in the direction of adopting NR as a standalone representation, by considering the possibility of performing downstream tasks by processing directly the NR weights. We first show that deep neural networks can be compressed into compact latent codes. Then, we show how this technique can be exploited to perform deep learning on implicit neural representations (INR) of 3D shapes, by only looking at the weights of the networks.