PURE > Dspace - Faculdade de Ciências
URI permanente para esta coleção:
Navegar
Entradas recentes
- Functional analysis of Chestnut and Phytophthora genes involved in plant-pathogen interactionPublication . Francisco, Eduardo Marcelo Rodrigues; Faculty of Sciences; Serrazina, Susana Maria Traquete; Fernandes, Patrícia MoraisInk disease, caused by the oomycete Phytophthora cinnamomi Rands, is a form of root rot that affects many woody plants worldwide, including chestnut trees. This pathogen has characteristics, such as long-lasting and chemical-resistant oospores, which make it considerably hard to control. Both the European (Castanea sativa Mill.) and the American [Castanea dentata (Marshall) Borkh.] chestnut species are susceptible to P. cinnamomi, having been historically affected by this oomycete. In contrast, the Asian chestnut species Castanea crenata Sieb. and Zucc. is resistant. The expression of the putative resistance gene Cast_Gnk2-like in this species might be one of the factors contributing to defense. Taking advantage of recently published data regarding the functional validation of Cast_Gnk2-like as a relevant defense factor and P. cinnamomi transcriptome in resistant and susceptible chestnut species throughout the infection process, the goals of this work are: 1) assess Cast_Gnk2-like overexpression from transformed C. sativa embryo clumps; 2) analyze the effect of Cast_Gnk2-like overexpression of transformed C. dentata on P. cinnamomi gene expression during the infection process; 3) analyze P. cinnamomi proteins using in silico tools to complement our understanding of their function and properties. We found that 3 out of 6 different transformed C. sativa lines had high enough significant overexpression and were selected for plant regeneration and further studies. Due to limitations of the study, no conclusions could be taken from the C. dentata inoculation assays, but in silico analysis results revealed many insights regarding the properties of the selected P. cinnamomi proteins.
- Modelação de sinistros de uma cobertura no ramo Multirriscos HabitaçãoPublication . Lourenço, Sofia Pedro Maciel; Faculdade de Ciências; Sousa, Lisete Maria Ribeiro deAs part of the development of a new tariff for multi-risk home insurance, the main objective of this project is to model the coverage of glass break. To this end, statistical modeling techniques are used, focusing on the use of Generalized Linear Models (GLM), in order to estimate both the frequency of occurrences (probability of an accident occurring within a period of time) and the average cost of claims associated with this coverage. The purpose of the study is to obtain a pure technical and actuarially fair premium, which allows to support the construction of a tariff appropriate to the current reality of the market and the observed accident rate. The methodology used is based on the application of the Emblem software, a tool widely used in the insurance sector, which allows the adjustment of GLM with different distributions, in an intuitive way and oriented to insurance pricing. In the case of frequency, a Poisson distribution is considered, suitable for modeling discrete and non-negative variables, such as the number of claims per policy. For the average cost, the Gamma distribution is used, appropriate for modeling continuous and positive values with a distribution skewed to the right, a common characteristic of claims costs. The analysis includes the segmentation of the portfolio based on relevant tariff variables, such as geographical location, housing typology, insured value, among others, allowing to capture the risk inherent to different customer profiles. The models are assessed on the basis of statistical criteria of quality of fit, as well as for their technical consistency and interpretability. The results obtained demonstrate the existence of significant variability in both frequency and average cost between different portfolio segments, reinforcing the importance of more granular and riskadjusted pricing. By combining the two models – frequency and average cost – it is possible to calculate the pure premium by segment, thus contributing to a more robust, sustainable and technically based tariff structure. This project intends, not only to directly support the development of a new tariff, but also to highlight the value of statistical modelling as an essential tool in the insurance pricing process, promoting fairer, more competitive practices aligned with the reality of risk.
- UterineExplorer(Plus) : toolbox for uterine eletromyography processing and visualizationPublication . Jorge, Cláudio José Marques; Faculty of Sciences; Department of Physics; Conceição, Raquel Cruz da; Batista, Arnaldo G.Preterm labor remains a major public health challenge and one of the leading causes of neonatal morbidity and mortality. The electrohysterogram (EHG), a non-invasive recording of uterine electrical activity, is considered a promising biomarker for predicting labor and preterm birth. However, the scarcity of robust and user-friendly software tools for multichannel EHG analysis has limited its wider adoption in both research and clinical practice. This dissertation presents the development of UterineExplorerPLUS (UExPLUS), a MATLAB-based platform designed to process, analyze, and visualize EHG signals. Migrating from the outdated GUIDE system to AppDesigner, UExPLUS introduces a modular and sustainable structure that integrates a wide range of methods within a unified environment. These include spectral and time-frequency analyses (Welch, Continuous Wavelet Transform, Empirical Mode Decomposition), nonlinear complexity measures (Approximate Entropy, Sample Entropy, Fuzzy Entropy, Conditional Entropy), clustering algorithms, and adaptive filtering. A key contribution is the implementation of a robust Alvarez wave detector, based on strict spectral criteria and concatenation of waves less than three seconds apart, while filtering maternal respiration artefacts. Other innovations include improved pacemaker estimation, intrauterine pressure (IUP) analysis, and maternal respiration processing. UExPLUS also enables multichannel visualization through scalograms and spectrograms, supporting the estimation of instantaneous frequency, bandwidth, and energy distributions. Validation with public datasets, such as the Icelandic 16-channel database, confirmed the reliability and versatility of the platform. Results demonstrate that UExPLUS provides accurate computation, clear visualization, and reproducible storage of multiple uterine activity parameters. This work contributes to computational obstetrics by offering an integrated and extensible toolbox for EHG analysis, paving the way for future clinical translation and supporting the early detection of labor complications, particularly preterm birth.
- Acompanhamento e implementação de um produto na sua construção e também no pós vendaPublication . Narciso, Diogo Miguel Jerónimo; Faculdade de Ciências; Gomes, João José Ferreira; Palma, Ana RitaThe implementation and monitoring of products in the insurance sector are processes that are considered fundamental to guaranteeing that the offer meets market needs, regulatory compliance and technical sustainability. This project report analyses, in an integrated way, the essential stages of this process, from conception and feasibility analysis, through technical and pricing definition, to post-launch monitoring, with the idea of the importance of ongoing management for the commercial and technical success of an insurance product. The methodology included a review of the literature and regulatory framework, as well as a detailed description of the implementation phases of an insurance product. To illustrate the practical application of these stages, the case of GAP insurance was analysed using real production and claims data. Statistical techniques were applied in the R software, including significance tests and multiple linear regression modelling, allowing variables with a relevant impact on the cost of claims to be identified. The results show that factors such as the cylinder ability of the vehicle, the type of fuel, the nature of the event and the status of the claim significantly influence the average costs, while seasonal and socio-demographic variables have minor impact. This evidence reinforces the importance of a structured technical approach and analytical tools for checking products, making it possible to adjust tariffs, improve internal processes and improve risk management. It can be concluded that the method presented can be applied to diverse types of insurance and is a good practice for ensuring competitiveness and efficiency in the sector. The use of predictive models and continuous monitoring systems is a strategic opportunity for the future of the technical management of insurance products.
- Testes de homogeneidade e heterogeneidade para os modelos de Loss Given Default (LGD)Publication . Amorim, Beatriz Santos de; Faculdade de Ciências; Alpuim, Maria Teresa dos Santos Hall de Agorreta de; Silva, Joana CostaThis project aimed to evaluate the segmentation of Loss Given Default (LGD) models, specifically their statistical robustness. The methodology applied was based on two fundamental principles: ensuring homogeneity within the final segments and ensuring heterogeneity between the final segments, in order to guarantee that the final segments are consistent in terms of customer differentiation and risk profile. To ensure homogeneity, it was established that segments with a standard deviation greater than 100% would be considered non-homogeneous, recommending further analysis and a possible revision of that segment. This analysis considered a statistical criterion whereby a division would only occur if the risk driver presented a p-value equal to or less than 5%. It is also important to mention that any possible disaggregation of segments must comply with the predefined definitions for each final segment, such as a minimum number of observations of 800 in that segment. Regarding heterogeneity, the objective was to identify segments with identical risk profiles for which there is no reason to remain disaggregated. Initially, the variance of each group of segments under analysis was examined, using the F-test to assess the equality of variances, in order to apply the most appropriate statistical test. When the two segments analyzed had equal variances, the T-test was applied to compare the means; otherwise, Welch's test was used, suitable for unequal variances. Finally, the discriminatory power was analyzed using the gAUC (generalized Area Under the Curve) performance metric of the proposed model, and it was compared with the model currently in use at the institution. The results show that the segmentation obtained with this project demonstrates better performance when compared to the segmentation currently utilized.
- Modelação de sinistros de custo elevado no ramo multirriscos habitaçãoPublication . Martins, Marta Maria Russo; Faculdade de Ciências; Diamantino, Maria Fernanda NunesThe insurance market is highly competitive, requiring insurers to continuously evolve their risk rating methodologies to ensure a balance between profitability and customer appeal. Traditional pricing approaches, often based on general averages, can lead to distortions in cost allocation, ultimately affecting the competitiveness of companies’. In this context, predictive models, such as logistic regression, as appliedin this study, play a strategic role in enhancing the accuracy of risk estimation, enabling fairer differentiation between contracts and a stronger competitive position in the market. This study aims to develop two predictive models based on the application of logistic regression, with the goal of estimating the probability of the occurrence of large claimsinnon-life insurance policies (Household). Two distinct models were developed, each focused on a specific coverage: water damage, with emphasis on the building itself, and electrical risks, with emphasis on the contents. Currently, high-cost claims are distributed uniformly across all contracts, regardless of the actual likelihood of occurrence, resulting in a less equitable cost distribution. The proposed model seeks to identify customer segments with a higher propensity for generating large claims, thereby enabling a more equitable and risk-adjusted pricing strategy for each contract. To achieve this, a response variable was defined to represent the proportion of high-cost claims, weighted by the total number of claims, ensuring a fairer and more efficient approach. The data used in this study were provided by an insurance company and processed to ensure confidentiality. The dataset comprises information from five complete financial years, spanning from 2019 to 2023, for the aforementioned coverages. The findings of this study aim to contribute to the refinement of technical pricing criteria, promoting a fairer cost allocation and a more accurate risk assessment
- Bioinformatics toolbox for comparative clustering evaluation of Whole-Genome Sequencing (WGS) pipelines for bacteria routine surveillancePublication . Pereira, Joana Vanessa Gomes; Faculty of Sciences; Mixão, Verónica de Pinho; Couto, Francisco José MoreiraWhole-Genome Sequencing (WGS) provides higher resolution than traditional typing to distinguish closely related isolates. As result, disease surveillance increasingly adopts WGS, with international agencies recommending its use in reference laboratories. However, the heterogeneity of workflows and unequal resources raise concerns about inter-laboratory result comparability and, consequently, data sharing and communication. To address these issues, this thesis project developed EvalTree, a Python-based command-line tool to compare clustering results from two typing solutions, including traditional and genome-scale approaches, assessing their congruence at all possible resolution levels. EvalTree accepts two input folders or clustering files, processes them, and produces multiple outputs, including an user-friendly HTML report. When a folder generated by ReporTree, a tool to identify genetic clusters at all possible distance thresholds, is provided as input, EvalTree enables not only the inter-pipeline clustering comparison, but also detection of stable clustering regions, cluster characterization using metadata, and assessment of outbreak signal overlap. EvalTree was validated and benchmarked using a large (2946 isolates) and diverse dataset of Salmonella enterica, showing it accurately reproduces a recently published large-scale evaluation of inter-pipeline congruence at the European level. Its running time was mainly affected by dataset diversity rather than size. To further demonstrate its applicability, EvalTree supported the implementation of the S. enterica genomic surveillance pipeline at the Portuguese National Institute of Health (INSA), by comparing its performance with that of the European Food Safety Authority (EFSA), revealing high cluster congruence and similar resolution power. In summary, EvalTree is a novel bioinformatics tool (available through conda installation) that offers a practical, flexible solution to evaluate cluster congruence between the pipelines of different laboratories, supporting inter-laboratory communication in a One Health framework. It also promotes the long-term sustainability of any pipeline by enabling informed decision-making throughout its life-cycle (e.g., evaluating software updates).
- mobileINFO seamless ticketing SDK and AppPublication . Rato, José Pedro Rijo das Neves Calcinha; Faculty of Sciences; Department of Informatics; Lourenço, Carlos Eduardo Ramos dos SantosPublic transportation increasingly uses smart card-based tickets as an alternative to paper tickets. As smartphones offer more capabilities and are widely adopted by consumers, Card4B - Systems S. A. has provided another alternative by developing an Android SDK that enables the digital purchase and validation of tickets. A mobile ticketing system poses several advantages, such as remote payment access, queue avoidance, and lack of need for cash handling. In the context of this project, it also helps address any potential issues regarding the complexity of transport networks to encourage the adoption of public transportation. The developed solution allows the use of public transportation without knowing the fare rules. An Android application has been published that allows journeys to be seamlessly tracked. This approach follows a post-paid billing system, where the service cost is charged at the end of the month and is based on the passenger’s usage. With the goal of entering the iOS market, multiplatform technologies were explored to build a single SDK, targeting different platforms to leverage reusability and maintainability instead of building a native equivalent SDK. This report details the transition from a platform-dependent SDK to a multiplatform SDK. The transition was made using the KMM framework. The framework proved to be practical and capable of supporting the development of the project as it was possible to build an SDK that could be used by native Android and iOS applications.
- Metodologia de avaliação do poder discriminatório no âmbito de low default portfolios (LDP)Publication . Leira, Carolina Lopes; Faculdade de Ciências; Alpuim, Maria Teresa dos Santos Hall de Agorreta de; Correia, Manuel Guilherme Laranjeira Pedrosa MartinsOne of the most commonly used techniques to determine the discriminatory power of a PD model is the Gini coefficient, a metric that measures the absolute difference in the distribution of customers in the portfolio/sample by risk grade. However, when Gini is applied to portfolios with a small number of customers and/or few high-risk customers, i.e. Low Default Portefolios (LDP), it may not be as reliable, creating the need to determine a confidence interval associated with the Gini. This work aims to address this issue and identify methods for estimating the uncertainty of the Gini, in order to provide more information on the performance of LDPs. Several methods for determining Gini uncertainty are explored, namely Bootstrap, Mann-Whitney, and F-Gini, applied to randomly generated samples and real portfolios from a Bank. Additionally, another method inspired by F-Gini was developed: MS-Gini. The study concludes that the most reliable methods for determining Gini uncertainty, among those analyzed, are Mann-Whitney and F-Gini, as they accurately describe the sensitivity of Gini in LDPs, increasing Gini uncertainty in portfolios with a small number of customers, few high-risk customers, or poor customer distribution. The F-Gini method can be applied to any portfolio; however, it requires some computation time. The Mann-Whitney method has only one restriction: it cannot be applied to portfolios with only one bad client. The Bootstrap method proved inadequate for calculating reliable uncertainty, and the MS-Gini method shows a bias in Gini uncertainty for samples with a small number of clients and many bad clients, resulting from the method’s construction itself, making it unreliable in these cases.
- BioSoundScape: Sonificação de BiossinaisPublication . Margarido, Carolina Rosa Nobre; Faculdade de Ciências; Departamento de Física ; Santos, Nuno Manuel Garcia dos; Gamboa, HugoIn today’s world, marked by rapid technological advances and the rise of artificial intelligence, traditionally valued skills such as memorization, prediction, and pattern recognition are losing relevance. For humans to thrive in the age of artificial intelligence, it is crucial not to replicate its behavior, but to use it as an ally to human uniqueness. In this scenario, it becomes urgent to explore new forms of teaching that emphasize creativity, emotion, and active student participation. This work investigates the potential of music personalization from biosignals as an educational tool, seeking to answer the central question: How can the personalization of music based on biosignals be used to teach concepts of physiology and improve users’ perception of their own body? To address this question, the BioSoundScape project was developed, transforming participants’ heartbeats into interactive musical experiences. These experiences taught secondary school students not only about the human body in general but also about their own physiology, fostering self-knowledge and self-regulation. Three experiences were created: What Do I Sound Like, Our Song, and Hearts in Sync, and presented at the Expo FCT event. Validation combined observation, surveys, and comparison with a control version based on synthesized sound. Results showed that most participants considered the experiences “very appealing,” reporting they had learned concepts such as heart rate, variability, and the effects of breathing and exercise. Moreover, adolescents preferred musical experiences aesthetically closer to the genres they usually listen to rather than abstract sonifications, confirming the importance of the aesthetic dimension in pedagogical success. It was concluded that the personalization of music based on biosignals is an innovative and effective approach to learning about the human body and physiology, paving the way to reinvent education by integrating it with art, emotion, and technology, aligned with the challenges of the digital age.