A carregar...
Projeto de investigação
Sem título
Financiador
Autores
Publicações
Map-based interaction with trajectory data
Publication . Gonçalves, Tiago José Lopes; Afonso, Ana Paula Pereira, 1965-; Martins, Bruno Emanuel da Graça, 1975-
With the increasing popularity of location based services and mobile tracking technologies, the collection of large amounts of spatio-temporal data became an increasingly common, easier, and more reliable task. In turn, this has emphasized the possibility of analysing georeferenced information, particularly associated with human trajectory data, to identify and understand movement patterns and activities, ultimately, supporting decision making in various contexts. In order to properly analyse and understand the spatio-temporal and the thematic properties associated with these data, adequate visualization techniques are needed. Due to the spatial properties of trajectories, map-based techniques, such as 2D static maps or 3D space-time cubes (STCs) are considered as essential tools for their visualization. However, despite the increasing number of visualization systems, the study regarding their usability, alongside the role of the human user, sometimes with a limited background in data visualization and analysis, are often neglected. In addition to the somewhat disperse, and sometimes even contradictory, results in the literature, these factors, ultimately, emphasize the lack of knowledge to support the choice of particular visualizations, and their design, in different types of tasks. This dissertation addresses these issues through three main sets of contributions, focusing on inexperienced users, in terms of data visualization and analysis: i) the characterization of the dis/advantages of existing map-based techniques (2D static maps and STCs), depending on the types of visual analysis tasks and the focus of the analysis; ii) the improvement of existing visualization techniques, either through the inclusion of additional spatial cues within the STC, or combining both types of techniques in various ways; and iii) the identification of design guidelines for trajectory data visualization, describing various considerations/criteria for the selection of different map-based visualization techniques and their possible interactive features.
Understanding ALS patients using Semantic Similarity
Publication . Teixeira, David Carriço; Pesquita, Cátia,1980-; Madeira, Sara Alexandra Cordeiro
As técnicas clássicas de prospecção de dados têm dificuldades a lidar com dados biomédicos não estruturados/ semiestruturados, pois estes contêm um significado semântico profundamente enraizado em palavras e frases que não é detectado através da extracção e análise diretas de recursos. Uma maneira de formalmente contextualizar dados é anotá-los com ontologias biomédicas e usar semelhança semântica sobre essas anotações para encontrar relações ocultas entre instâncias de dados. Deste modo, se os dados puderem ser enriquecidos com conhecimento externo, uma prospecção mais informada poderá, em princípio, retornar resultados mais precisos. Este projeto abordou este desafio desenvolvendo uma metodologia para analisar registos médicos de pacientes por meio da integração com recursos e software semânticos. Uma pipeline de três etapas cria uma rede semântica de ontologias que garante cobertura semântica sobre os dados alvo, calcula a semelhança semântica entre pacientes com a aplicação SML (Semantic Measures Library), e agrupa pacientes usando algoritmos de clustering do módulo Scikit-Learn do Python. Além disso, foi desenvolvida uma ferramenta para elaborar uma descrição resumida do conteúdo semântico de um agrupamento, destacando os seus elementos mais relevantes. Estes métodos foram avaliados usando um conjunto de dados de 1376 pacientes com esclerose lateral aiotrófica (ELA), possuindo uma forte componente textual e uma ampla heterogeneidade de sintomas entre pacientes. Os grupos de pacientes obtidos foram comparados, juntamente com uma baseline não-semântica, com grupos ground-truth de pacientes derivados das suas taxas de progressão de ELA. Foi demonstrado que a eficácia da metodologia proposta era fortemente dependente do número e da qualidade das anotações, mas também que os dados disponíveis não eram suficientes para detectar grupos de progressão. Apesar disso, as descrições de agrupamentos foram aplicadas com êxito em todas as abordagens, e forneceram informações úteis que evidenciaram pontos em comum entre o conteúdo semântico dos agrupamentos teste e da ground-truth. Por fim, esta metodologia pode ser generalizada para quaisquer entidades biomédicas que podem ser anotadas semanticamente com ontologias existentes.
Extracting biomedical relations from biomedical literature
Publication . Maldonado, Tânia Sofia Guerreiro; Couto, Francisco José Moreira
A ciência, e em especial o ramo biomédico, testemunham hoje um crescimento de conhecimento a uma taxa que clínicos, cientistas e investigadores têm dificuldade em acompanhar. Factos científicos espalhados por diferentes tipos de publicações, a riqueza de menções etiológicas, mecanismos moleculares, pontos anatómicos e outras terminologias biomédicas que não se encontram uniformes ao longo das várias publicações, para além de outros constrangimentos, encorajaram a aplicação de métodos de text mining ao processo de revisão sistemática. Este trabalho pretende testar o impacto positivo que as ferramentas de text mining juntamente com vocabulários controlados (enquanto forma de organização de conhecimento, para auxílio num posterior momento de recolha de informação) têm no processo de revisão sistemática, através de um sistema capaz de criar um modelo de classificação cujo treino é baseado num vocabulário controlado (MeSH), que pode ser aplicado a uma panóplia de literatura biomédica. Para esse propósito, este projeto divide-se em duas tarefas distintas: a criação de um sistema, constituído por uma ferramenta que pesquisa a base de dados PubMed por artigos científicos e os grava de acordo com etiquetas pré-definidas, e outra ferramenta que classifica um conjunto de artigos; e a análise dos resultados obtidos pelo sistema criado, quando aplicado a dois casos práticos diferentes. O sistema foi avaliado através de uma série de testes, com recurso a datasets cuja classificação era conhecida, permitindo a confirmação dos resultados obtidos. Posteriormente, o sistema foi testado com recurso a dois datasets independentes, manualmente curados por investigadores cuja área de investigação se relaciona com os dados. Esta forma de avaliação atingiu, por exemplo, resultados de precisão cujos valores oscilam entre os 68% e os 81%. Os resultados obtidos dão ênfase ao uso das tecnologias e ferramentas de text mining em conjunto com vocabulários controlados, como é o caso do MeSH, como forma de criação de pesquisas mais complexas e dinâmicas que permitam melhorar os resultados de problemas de classificação, como são aqueles que este trabalho retrata.
IsomiR_Window : a system for analyzing small‑RNA‑seq data in an integrative and user‑friendly manner
Publication . Vasconcelos, Ana M.; Carmo, Maria Beatriz; Ferreira, Beatriz; Viegas, Inês; Gama‑Carvalho, Margarida; Ferreira, António; Amaral, Andreia J.
Background: IsomiRs are miRNA variants that vary in length and/or sequence when
compared to their canonical forms. These variants display differences in length and/or
sequence, including additions or deletions of one or more nucleotides (nts) at the 5′
and/or 3′ end, internal editings or untemplated 3′ end additions. Most available tools
for small RNA-seq data analysis do not allow the identification of isomiRs and often
require advanced knowledge of bioinformatics. To overcome this, we have developed
IsomiR Window, a platform that supports the systematic identification, quantification
and functional exploration of isomiR expression in small RNA-seq datasets, accessible
to users with no computational skills.
Methods: IsomiR Window enables the discovery of isomiRs and identification of all
annotated non-coding RNAs in RNA-seq datasets from animals and plants. It comprises
two main components: the IsomiR Window pipeline for data processing; and
the IsomiR Window Browser interface. It integrates over ten third-party softwares for
the analysis of small-RNA-seq data and holds a new algorithm that allows the detection
of all possible types of isomiRs. These include 3′ and 5′end isomiRs, 3′ end tailings,
isomiRs with single nucleotide polymorphisms (SNPs) or potential RNA editings, as well
as all possible fuzzy combinations. IsomiR Window includes all required databases for
analysis and annotation, and is freely distributed as a Linux virtual machine, including
all required software.
Results: IsomiR Window processes several datasets in an automated manner, without
restrictions of input file size. It generates high quality interactive figures and tables
which can be exported into different formats. The performance of isomiR detection
and quantification was assessed using simulated small-RNA-seq data. For correctly
mapped reads, it identified different types of isomiRs with high confidence and 100%
accuracy. The analysis of a small RNA-seq data from Basal Cell Carcinomas (BCCs)
using isomiR Window confirmed that miR-183-5p is up-regulated in Nodular BCCs, but
revealed that this effect was predominantly due to a novel 5′end variant. This variant
displays a different seed region motif and 1756 isoform-exclusive mRNA targets that
are significantly associated with disease pathways, underscoring the biological relevance
of isomiR-focused analysis. IsomiR Window is available at https ://isomi r.fc.ul.pt/.
Byzantine state machine replication for the masses
Publication . Sousa, João; Bessani, Alysson Neves, 1978-
The state machine replication technique is a popular approach for building Byzantine fault-tolerant services. However, despite the widespread adoption of this paradigm for crash fault-tolerant systems, there are still few examples of this paradigm for real Byzantine fault-tolerant systems. Our view of this situation is that there is a lack of robust implementations of Byzantine fault-tolerant state machine replication middleware, and that the performance penalty is too high, specially for geo-replication. These hindrances are tightly coupled to the distributed protocols used for enforcing such resilience. This thesis has the objective of finding methodologies for enhancing robustness and performance of state machine replication systems. The first contribution is Mod-SMaRt, a modular protocol that preserves optimal latency in terms of the communications steps exchanged among processes. By being a modular protocol, it becomes simpler to validate and implement, thus resulting in greater robustness; by also preserving optimal message-exchanges among processes, the protocol is capable of delivering desirable performance. The second contribution is concerned with implementing Mod-SMaRt into BFTSMART, a reliable and high-performance codebase that was maintained and improved over the entire course of the PhD that offers multicore-awareness, reconfiguration support, and a flexible API. The third contribution presents WHEAT, a protocol derived from Mod-SMaRt that uses optimizations shown to be effective in reducing latency via a practical evaluation conducted in a geo distributed environment. We additionally conducted an evaluation of both BFT-SMART and WHEAT applied to a relational database middleware and an ordering service for a permissioned blockchain platform. These evaluations revealed encouraging results for both systems and validated our work conducted in the geo-distributed context.
Unidades organizacionais
Descrição
Palavras-chave
Contribuidores
Financiadores
Entidade financiadora
Fundação para a Ciência e a Tecnologia
Programa de financiamento
5876
Número da atribuição
UID/CEC/00408/2013
