Analyzing the dependence structure of microarray data: a copula–based approach

The main aim of this Ph.D. dissertation is the study of clustering dependent data by means of copula functions with particular emphasis on microarray data. Copula functions are a popular multivariate modeling tool in each field where the multivariate dependence is of great interest and their use in clustering has not been still investigated. The first part of this work contains the review of the literature of clustering methods, copula functions and microarray experiments. The attention focuses on the K–means (Hartigan, 1975; Hartigan and Wong, 1979), the hierarchical (Everitt, 1974) and the model–based (Fraley and Raftery, 1998, 1999, 2000, 2007) clustering techniques because their performance is compared. Then, the probabilistic interpretation of the Sklar’s theorem (Sklar’s, 1959), the estimation methods for copulas like the Inference for Margins (Joe and Xu, 1996) and the Archimedean and Elliptical copula families are presented. In the end, applications of clustering methods and copulas to the genetic and microarray experiments are highlighted. The second part contains the original contribution proposed. A simulation study is performed in order to evaluate the performance of the K–means and the hierarchical bottom–up clustering methods in identifying clusters according to the dependence structure of the data generating process. Different simulations are performed by varying different conditions (e.g., the kind of margins (distinct, overlapping and nested) and the value of the dependence parameter ) and the results are evaluated by means of different measures of performance. In light of the simulation results and of the limits of the two investigated clustering methods, a new clustering algorithm based on copula functions (‘CoClust’ in brief) is proposed. The basic idea, the iterative procedure of the CoClust and the description of the written R functions with their output are given. The CoClust algorithm is tested on simulated data (by varying the number of clusters, the copula models, the dependence parameter value and the degree of overlap of margins) and is compared with the performance of model–based clustering by using different measures of performance, like the percentage of well–identified number of clusters and the not rejection percentage of H0 on . It is shown that the CoClust algorithm allows to overcome all observed limits of the other investigated clustering techniques and is able to identify clusters according to the dependence structure of the data independently of the degree of overlap of margins and the strength of the dependence. The CoClust uses a criterion based on the maximized log–likelihood function of the copula and can virtually account for any possible dependence relationship between observations. Many peculiar characteristics are shown for the CoClust, e.g. its capability of identifying the true number of clusters and the fact that it does not require a starting classification. Finally, the CoClust algorithm is applied to the real microarray data of Hedenfalk et al. (2001) both to the gene expressions observed in three different cancer samples and to the columns (tumor samples) of the whole data matrix.

Analyzing convergence of Agricultural and Rural Areas in the Enlarged EU: The case of Hungary

The fall of the Berlin Wall opened the way for a reform path – the transition process – which accompanied ten former Socialist countries in Central and South Eastern Europe to knock at the EU doors. By the way, at the time of the EU membership several economic and structural weaknesses remained. A tendency towards convergence between the new Member States (NMS) and the EU average income level emerged, together with a spread of inequality at the sub-regional level, mainly driven by the backwardness of the agricultural and rural areas. Several progresses were made in evaluating the policies for rural areas, but a shared definition of rurality is still missing. Numerous indicators were calculated for assessing the effectiveness of the Common Agricultural Policy and Rural Development Policy. Previous analysis on the Central and Eastern European countries found that the characteristics of the most backward areas were insufficiently addressed by the policies enacted; the low data availability and accountability at a sub-regional level, and the deficiencies in institutional planning and implementation represented an obstacle for targeting policies and payments. The next pages aim at providing a basis for understanding the connections between the peculiarities of the transition process, the current development performance of NMS and the EU role, with particular attention to the agricultural and rural areas. Applying a mixed methodological approach (multivariate statistics, non-parametric methods, spatial econometrics), this study contributes to the identification of rural areas and to the analysis of the changes occurred during the EU membership in Hungary, assessing the effect of CAP introduction and its contribution to the convergence of the Hungarian agricultural and rural. The author believes that more targeted – and therefore efficient – policies for agricultural and rural areas require a deeper knowledge of their structural and dynamic characteristics.

Analyzing urban green adaptation opportunities: concepts, approaches, & strategies for existing neighborhoods

Over the last decades, the growing evidence of human-caused climate change has raised awareness of the consequences of exceeding global temperature by 2˚C. This awareness has led to a contemporary approach to the conceptualization and management of green adaptation policies in spatial planning. This thesis aims to develop a comprehensive methodology for assessing the adaptability of existing neighborhoods to green strategies. The reliability of the proposed method is examined in the cities of Bologna and Imola and proved to be applicable in other geoghraphical locations. This thesis integrates three key themes of conceptual and implementation principles for urban green adaptation. This thesis initially defines methods for narrowing uncertainties in urban planning energy forecasting modeling by exploring the roles of integrated energy planning. The second is by exploring green retrofitting strategies in building, this thesis examines the effects of various energy-saving factors in roofing scenarios including a green roof, rooftop greenhouse, and insolated roof. Lastly, this thesis analyzes green strategies in urban spaces to enhance thermal comfort through facing urban heat exposure related to urban heat island effects. The roles of integrated energy policies and green strategic thinking are discussed to highlight various aspects of green adaptation on the neighborhood scale. This thesis develops approaches by which cities can face the challenges of current green urban planning and connect the conceptual and practical aspects of green spatial planning. Another point that this thesis highlight is that due to the interdependency of individuals and places, it is difficult to assure whether all the adaptation policies on a large scale are enhancing the resiliency of the neighborhood or they are simply shuffling the vulnerability through the individuals and places. Besides, it asserts that neglecting to reflect on these reallocations of the effects generates serious complications, and will result in long-term dysfunctional consequences.

Interpretable Convolutional Neural Networks for Decoding and Analyzing Neural Time Series Data

Machine learning is widely adopted to decode multi-variate neural time series, including electroencephalographic (EEG) and single-cell recordings. Recent solutions based on deep learning (DL) outperformed traditional decoders by automatically extracting relevant discriminative features from raw or minimally pre-processed signals. Convolutional Neural Networks (CNNs) have been successfully applied to EEG and are the most common DL-based EEG decoders in the state-of-the-art (SOA). However, the current research is affected by some limitations. SOA CNNs for EEG decoding usually exploit deep and heavy structures with the risk of overfitting small datasets, and architectures are often defined empirically. Furthermore, CNNs are mainly validated by designing within-subject decoders. Crucially, the automatically learned features mainly remain unexplored; conversely, interpreting these features may be of great value to use decoders also as analysis tools, highlighting neural signatures underlying the different decoded brain or behavioral states in a data-driven way. Lastly, SOA DL-based algorithms used to decode single-cell recordings rely on more complex, slower to train and less interpretable networks than CNNs, and the use of CNNs with these signals has not been investigated. This PhD research addresses the previous limitations, with reference to P300 and motor decoding from EEG, and motor decoding from single-neuron activity. CNNs were designed light, compact, and interpretable. Moreover, multiple training strategies were adopted, including transfer learning, which could reduce training times promoting the application of CNNs in practice. Furthermore, CNN-based EEG analyses were proposed to study neural features in the spatial, temporal and frequency domains, and proved to better highlight and enhance relevant neural features related to P300 and motor states than canonical EEG analyses. Remarkably, these analyses could be used, in perspective, to design novel EEG biomarkers for neurological or neurodevelopmental disorders. Lastly, CNNs were developed to decode single-neuron activity, providing a better compromise between performance and model complexity.

Integrating the age factor in designing industrial environments and workstations in the workforce ageing era

As people spend a third of their lives at work and, in most cases, indoors, the work environment assumes crucial importance. The continuous and dynamic interaction between people and the working environment surrounding them produces physiological and psychological effects on operators. Recognizing the substantial impact of comfort and well-being on employee satisfaction and job performance, the literature underscores the need for industries to implement indoor environment control strategies to ensure long-term success and profitability. However, managing physical risks (i.e., ergonomic and microclimate) in industrial environments is often constrained by production and energy requirements. In the food processing industry, for example, the safety of perishable products dictates storage temperatures that do not allow for operator comfort. Conversely, warehouses dedicated to non-perishable products often lack cooling systems to limit energy expenditure, reaching high temperatures in the summer period. Moreover, exceptional events, like the COVID-19 pandemic, introduce new constraints, with recommendations impacting thermal stress and respiratory health. Furthermore, the thesis highlights how workers' variables, particularly the aging process, reduce tolerance to environmental stresses. Consequently, prolonged exposure to environmental stress conditions at work results in cardiovascular disease and musculoskeletal disorders. In response to the global trend of an aging workforce, the thesis bridges a literature gap by proposing methods and models that integrate the age factor into comfort assessment. It aims to present technical and technological solutions to mitigate microclimate risks in industrial environments, ultimately seeking innovative ways to enhance the aging workforce's comfort, performance, experience, and skills. The research outlines a logical-conceptual scheme with three main areas of focus: analyzing factors influencing the work environment, recognizing constraints to worker comfort, and designing solutions. The results significantly contribute to science by laying the foundation for new research in worker health and safety in an ageing working population's extremely current industrial context.