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.

Populism as a battlefield Populist actors and their representation on social and legacy media during the european parliament elections 2019 in Finland, Italy and the Netherlands

The objective of the present research is to describe and explain populist actors and populism as a concept and their representation on social and legacy media during the 2019 EU elections in Finland, Italy and The Netherlands. This research tackles the topic of European populism in the context of political communication and its relation to both the legacy and digital media within the hybrid media system. Departing from the consideration that populism and populist rhetoric are challenging concepts to define, I suggest that they should be addressed and analyzed through the usage of a combination of methods and theoretical perspectives, namely Communication Studies, Corpus Linguistics, Political theory, Rhetoric and Corpus-Assisted Discourse Studies. This thesis considers data of different provenance. On the one hand, for the Legacy media part, newspapers articles were collected in the three countries under study from the 1st until the 31st of May 2019. Each country’s legacy system is represented by three different quality papers and the articles were collected according to a selection of keywords (European Union Elections and Populism in each of the three languages). On the other hand, the Digital media data takes into consideration Twitter tweets collected during the same timeframe based on particular country-specific hashtags and tweets by identified populist actors. In order to meet the objective of this study, three research questions are posed and the analysis leading to the results are exhaustively presented and further discussed. The results of this research provide valuable and novel insights on how populism as a theme and a concept is being portrayed in the context of the European elections both in legacy and digital media and political communication in general.

RICORDACI: Research on the Conservation, Restoration and Diagnosis of Cinematographic Films

This thesis is the result of the RICORDACI project, a three-year European-funded initiative involving the collaboration between the University of Bologna and the restoration laboratory of the Cineteca di Bologna, L'immagine Ritrovata, which aimed to develop innovative solutions and technologies for the preservation of cinematographic film heritage. In particular, this thesis presents new analytical methodologies to exploit two types of portable miniaturized Near Infrared spectrometers working in Diffuse Reflectance over the Short Wave Infrared (SWIR) range, to study the near infrared (NIR) spectral behavior of film base materials for an accurate, non-invasive and fast characterization of the polymer type; and for films with cellulose acetate supports, they can be employed as a diagnostic tool for monitoring the Degree of substitution (DS) affected by the loss of acetyl groups. The proposed methods offer non-invasive, fast, inexpensive and simple alternatives for the characterization and diagnosis of film bases to help the strategic planning and decision-making regarding storage, digitalization and intervention of film collections. Secondly, the thesis includes the evaluation of new green cleaning systems and solvents for the effective, fast and innocuous removal of undesired substances from degraded cinematographic films bases; these tests compared the efficiency of traditional systems and solvents against the new proposals. Firstly, the use of Deep Eutectic Solvent formulations for removing softened gelatin residues from cellulose nitrate bases; and secondly, the employment of green volatile solvents with different application methods, including the use of new electrospun nylon mats, for avoiding the dangerous use of friction for the removal of Triphenyl Phosphate blooms from the surface of cellulose acetate bases. The results obtained will help improving the efficiency of the interventions needed before the digitalization of historical cinematographic films and will pave the way for further investigation on the use of green solvents for cleaning polymeric heritage objects.

The European legal approach to Open Science and research data

This dissertation proposes an analysis of the governance of the European scientific research, focusing on the emergence of the Open Science paradigm: a new way of doing science, oriented towards the openness of every phase of the scientific research process, able to take full advantage of the digital ICTs. The emergence of this paradigm is relatively recent, but in the last years it has become increasingly relevant. The European institutions expressed a clear intention to embrace the Open Science paradigm (eg., think about the European Open Science Cloud, EOSC; or the establishment of the Horizon Europe programme). This dissertation provides a conceptual framework for the multiple interventions of the European institutions in the field of Open Science, addressing the major legal challenges of its implementation. The study investigates the notion of Open Science, proposing a definition that takes into account all its dimensions related to the human and fundamental rights framework in which Open Science is grounded. The inquiry addresses the legal challenges related to the openness of research data, in light of the European Open Data framework and the impact of the GDPR on the context of Open Science. The last part of the study is devoted to the infrastructural dimension of the Open Science paradigm, exploring the e-infrastructures. The focus is on a specific type of computational infrastructure: the High Performance Computing (HPC) facility. The adoption of HPC for research is analysed from the European perspective, investigating the EuroHPC project, and the local perspective, proposing the case study of the HPC facility of the University of Luxembourg, the ULHPC. This dissertation intends to underline the relevance of the legal coordination approach, between all actors and phases of the process, in order to develop and implement the Open Science paradigm, adhering to the underlying human and fundamental rights.

Deep learning and embeddings for problems of computational biology

The development of Next Generation Sequencing promotes Biology in the Big Data era. The ever-increasing gap between proteins with known sequences and those with a complete functional annotation requires computational methods for automatic structure and functional annotation. My research has been focusing on proteins and led so far to the development of three novel tools, DeepREx, E-SNPs&GO and ISPRED-SEQ, based on Machine and Deep Learning approaches. DeepREx computes the solvent exposure of residues in a protein chain. This problem is relevant for the definition of structural constraints regarding the possible folding of the protein. DeepREx exploits Long Short-Term Memory layers to capture residue-level interactions between positions distant in the sequence, achieving state-of-the-art performances. With DeepRex, I conducted a large-scale analysis investigating the relationship between solvent exposure of a residue and its probability to be pathogenic upon mutation. E-SNPs&GO predicts the pathogenicity of a Single Residue Variation. Variations occurring on a protein sequence can have different effects, possibly leading to the onset of diseases. E-SNPs&GO exploits protein embeddings generated by two novel Protein Language Models (PLMs), as well as a new way of representing functional information coming from the Gene Ontology. The method achieves state-of-the-art performances and is extremely time-efficient when compared to traditional approaches. ISPRED-SEQ predicts the presence of Protein-Protein Interaction sites in a protein sequence. Knowing how a protein interacts with other molecules is crucial for accurate functional characterization. ISPRED-SEQ exploits a convolutional layer to parse local context after embedding the protein sequence with two novel PLMs, greatly surpassing the current state-of-the-art. All methods are published in international journals and are available as user-friendly web servers. They have been developed keeping in mind standard guidelines for FAIRness (FAIR: Findable, Accessible, Interoperable, Reusable) and are integrated into the public collection of tools provided by ELIXIR, the European infrastructure for Bioinformatics.

Diet Effects on Activation and Maturation of Feed Control over the Gastrointestinal Defence Barrier in Piglets

Weaning is an important and complex step involving many stresses that interfere deeply with feed intake, gastro-intestinal tract (GIT) development and adaptation to the weaning diet in young pigs. The health of the pig at weaning, its nutrition in the immediate post-weaning period, and the physical, microbiological and psychological environment are all factors that interact to determine food intake and subsequent growth. GIT disorders, infections and diarrhoea increase at the time of weaning, in fact pathogens such as enterotoxigenic Escherichia coli (ETEC) are major causes of mucosal damage in post-weaning disease contributing to diarrhoea in suckling and post-weaned pigs. The European ban in 2006 put on antibiotic growth promoters (AGP) has stimulated research on the mechanisms of GIT disorders and on nutritional approaches for preventing or reducing such disturbances avoiding AGPs. Concerning these aspects here are presented five studies based on the interplay among nutrition, genomic, immunity and physiology with the aim to clarify some of these problematic issues around weaning period in piglets. The first three evaluate the effects of diets threonine or tryptophan enriched on gut defence and health as possible alternatives to AGP in the gut. The fourth is focused on the possible immunological function related with the development of the stomach. The fifth is a pilot study on the gastric sensing and orexygenic signal given by fasting or re-feeding conditions. Although some results are controversial, it appears that both tryptophan and threonine supplementation in weaning diets have a preventive role in E.coli PWD and favorable effects in the gut especially in relation to ETEC susceptible genotype. While the stomach is believed as almost aseptic organ, it shows an immune activity related with the mucosal maturation. Moreover it shows an orexygenic role of both oxyntic mucosa and pyloric mucosa, and its possible relation with nutrient sensing stimuli.