Logo do repositório
 
A carregar...
Logótipo do projeto
Projeto de investigação

Sem título

Autores

Publicações

Evolutionary online behaviour learning and adaptation in robotic systems
Publication . Silva, Fernando Goulart da; Correia, Luís Miguel Parreira e, 1959-; Christensen, Anders Lyhne
In this thesis, we study new ways to enable ecient online learning in autonomous robots. We employ a control synthesis methodology called evolutionary robotics, which emerged in the 1990s as a promising alternative to classic articial intelligence techniques and design methodologies for control systems. In online learning through evolution, henceforth called online evolution, an evolutionary algorithm is executed onboard each robot in order to create and continuously optimise its behavioural control logic. Each instance of the evolutionary algorithm executes without any external supervision or human intervention. Online evolution can thus automatically generate the articial intelligence that controls each robot, and creates the potential for long-term behaviour adaptation and learning: robots can continuously self adjust and learn new behaviours in response to, for example, changes in the task requirements or environmental conditions, and to faults in the sensors and/or actuators. Despite the potential for automatic behaviour learning, online evolution is not frequently employed for a number of reasons. First, online evolution typically requires several hours or days to synthesise solutions to a task. As a result, the approach has not yet been practically exploited in real-robot systems. Second, one common assumption in the eld is that online evolution enables continuous learning and adaptation to previously unforeseen circumstances. However, only a small number of ad-hoc experiments have been carried out in simulation. That is, the potential for online evolution to enable dynamic adaptation and learning has been largely left unstudied. The main goal of this thesis is to address some of the fundamental issues associated with online evolution to bring it closer to widespread adoption. Our research focuses on studying if and how to accelerate and increase the performance of online evolution. Our rst research contribution is a comprehensive presentation and analysis of Online Decentralised NeuroEvolution of Augmenting Topologies (odNEAT), an algorithm for online evolution of neural networkbased controllers in multirobot systems. odNEAT diers from more traditional approaches to online evolution because both the weighting parameters and the topological structure of neural networks are under evolutionary control. Our second research contribution focuses on investigating the dynamics of online evolution of controllers at two dierent levels. At the microscopic scale, we assess the dynamics of distinct neuronal models. At the macroscopic scale, we investigate the scalability properties of online evolution with respect to group size. The outcomes of the contribution are an assertion of the critical role of the controller evaluation policy, and an analysis of how the group size inuences task performance. In our third research contribution, we capitalise on the knowledge gained from the second study, and we introduce: (i) a racing approach that allows individual robots to cut short the evaluation of poor controllers, (ii) a population cloning approach that enables each individual robot to clone and transmit a varying number of high-performing controllers to other robots nearby, and (iii) online hyper-evolution (OHE), an unprecedented approach in evolutionary robotics with the capability to automatically construct algorithms for controller generation during task execution. To conclude, we validate our research in real robotic hardware, and we successfully demonstrate evolution of controllers to solve three classic evolutionary robotics tasks in a timely manner (one hour or less).
Effect of the microenvironment on alternative splicing in colorectal cells
Publication . Pereira, Joana FS; Jordan, Peter
The mucosa of our gastrointestinal tract forms a barrier that separates the gut lumen from the surrounding tissues. Perturbation of the barrier function characterizes inflammatory bowel diseases, which constitute a risk factor for colon cancer development, because an inflammatory microenvironment is a tumor-promoting condition that provides survival signals to which cancer cells respond with changes in their gene expression. One key regulatory mechanism is alternative splicing. One example is RAC1B, a RAC1 alternative splicing variant, that the host laboratory previously identified in a subset of BRAF-mutated colorectal tumors, and was found increased in samples from inflammatory bowel disease or following experimentally-induced acute colitis in a mouse model. The main goal of this work was to generate co-culture models to understand whether colorectal cancer cells respond to a pro-inflammatory microenvironment with changes in alternative splicing of RAC1B. For this, culture conditions for polarized 2D and 3D models were established as physiologically more relevant colon cell models. Caco-2 colorectal cancer cells were grown as polarized cells on porous membranes and then co-cultured with stromal cells: fibroblasts, monocytes and macrophages. RAC1B expression was analyzed in Caco-2 by qRT-PCR, Western blot and confocal fluorescence microscopy. Co-culture experiments revealed that the presence of fibroblasts and/or M1 macrophages induced a transient increase in RAC1B protein levels in the colorectal cells, accompanied by a loss of epithelial organization. Through the use of a human inflammation array, we were able to identify three cytokines in co-culture media that associated with increased RAC1B: IL-1β, IL-6 and IL-8. In addition, the analysis of cytokine-stimulated signaling pathways allowed to identify the phosphorylation of ERK 1/2 and STAT3 as a major response to the external signals. Overall, our data present evidence that a pro-inflammatory environment triggers an increase in RAC1B expression in colon epithelial cells. Since RAC1B was shown to sustain tumor cell survival and promote escape form oncogene-induced senescence, the data further strengthen the causal connection between inflammatory conditions and the development of colorectal cancer, and may indicate RAC1B as a potential biomarker.
Novel approaches to cooperative coevolution of heterogeneous multiagent systems
Publication . Gomes, Jorge Miguel Carvalho; Christensen, Anders Lyhne; Mariano, Pedro, 1975-; Correia, Luís Miguel Parreira e, 1959-
Heterogeneous multirobot systems are characterised by the morphological and/or behavioural heterogeneity of their constituent robots. These systems have a number of advantages over the more common homogeneous multirobot systems: they can leverage specialisation for increased efficiency, and they can solve tasks that are beyond the reach of any single type of robot, by combining the capabilities of different robots. Manually designing control for heterogeneous systems is a challenging endeavour, since the desired system behaviour has to be decomposed into behavioural rules for the individual robots, in such a way that the team as a whole cooperates and takes advantage of specialisation. Evolutionary robotics is a promising alternative that can be used to automate the synthesis of controllers for multirobot systems, but so far, research in the field has been mostly focused on homogeneous systems, such as swarm robotics systems. Cooperative coevolutionary algorithms (CCEAs) are a type of evolutionary algorithm that facilitate the evolution of control for heterogeneous systems, by working over a decomposition of the problem. In a typical CCEA application, each agent evolves in a separate population, with the evaluation of each agent depending on the cooperation with agents from the other coevolving populations. A CCEA is thus capable of projecting the large search space into multiple smaller, and more manageable, search spaces. Unfortunately, the use of cooperative coevolutionary algorithms is associated with a number of challenges. Previous works have shown that CCEAs are not necessarily attracted to the global optimum, but often converge to mediocre stable states; they can be inefficient when applied to large teams; and they have not yet been demonstrated in real robotic systems, nor in morphologically heterogeneous multirobot systems. In this thesis, we propose novel methods for overcoming the fundamental challenges in cooperative coevolutionary algorithms mentioned above, and study them in multirobot domains: we propose novelty-driven cooperative coevolution, in which premature convergence is avoided by encouraging behavioural novelty; and we propose Hyb-CCEA, an extension of CCEAs that places the team heterogeneity under evolutionary control, significantly improving its scalability with respect to the team size. These two approaches have in common that they take into account the exploration of the behaviour space by the evolutionary process. Besides relying on the fitness function for the evaluation of the candidate solutions, the evolutionary process analyses the behaviour of the evolving agents to improve the effectiveness of the evolutionary search. The ultimate goal of our research is to achieve general methods that can effectively synthesise controllers for heterogeneous multirobot systems, and therefore help to realise the full potential of this type of systems. To this end, we demonstrate the proposed approaches in a variety of multirobot domains used in previous works, and we study the application of CCEAs to new robotics domains, including a morphological heterogeneous system and a real robotic system.
How sex led to wine: Genomic and post-transcriptional differences associated to sex in Vitis vinifera sylvestris
Publication . Ramos, Miguel de Jesus Nunes Ferreira; Rocheta, Margarida Maria Pedro; Costa, Maria Manuela Ribeiro
The interest in the study of Vitis is closely associated with the importance of wine and the economic impact of this species. Within this duality, two subspecies gain relevant attention: Vitis vinifera vinifera (vinifera) is the subspecies most used in viticulture, contains essentially hermaphroditic plants and results from a long process of domestication; and Vitis vinifera sylvestris (sylvestris), an essentially dioecious subspecies (with only a few hermaphroditic individuals per population), which has not been domesticated. The comparative study of both subspecies is important, as it increases the general knowledge in relation to Vitis genus, namely, to clarify which genes are associated with its sexual determination. In order to deepen the knowledge regarding the sexual determination of Vitis, studies were performed in the areas of transcritomics, genomics and bioinformatics. One of the main conclusions of this work is the possible involvement of posttranscriptional regulators, such as micro RNAs, and RNA editing in nuclear transcripts, in a systematic and constant manner. The extensive validation of these results showed an association between these posttranscriptional players and the flower developmental stage, as well as the flower type (male or female), with consistent results in consecutive years analysed. It should also be emphasized that some of the processes studied, namely the RNA editing, can be transversal to other species and organs and so, the results found here can enhance knowledge in other systems
Study on the regulation of the expression of alternative protein isoforms involved in carcinogenesis
Publication . Pereira, Bruna Filipa Francisco; Loison, Luísa Romão,1963-; Candeias, Marco Marques
Em eucariotas, a passagem da informação genética contida no ácido desoxirribonucleico (DNA, do inglês deoxyribonucleic acid) para proteínas funcionais envolve uma série de processos que estão física e funcionalmente interligados. Tudo se inicia no núcleo da célula, com a transcrição da informação armazenada no DNA para moléculas de ácido ribonucleico mensageiro prematuro (pré-mRNA, do inglês premature ribonucleic acid). À medida que as moléculas de pré-mRNA vão sendo formadas, estas vão sofrendo modificações que visam a sua estabilização e preparação para as fases seguintes da expressão génica. A estrutura cap (guanina metilada, m7G) é adicionada na extremidade 5’ do pré-mRNA, intrões (regiões não codificantes) são removidos, com consequente junção dos exões (regiões codificantes), num processo designado splicing, e a extremidade 3’ do pré-mRNA sofre poliadenilação. Após este processamento, o mRNA maduro é transportado para o citoplasma, para que a tradução para proteína possa ocorrer nos ribossomas. De uma forma geral, esta inicia-se com a formação do complexo ternário, composto pelo fator de iniciação da tradução (eIF, do inglês eukaryotic translation initiation factor) 2 ligado a uma guanosina trifosfato (GTP, do inglês, guanosine triphosphate) e a uma molécula de RNA de transferência que transporta a metionina da primeira cadeia peptídica (Met–tRNAi, do inglês initiator transfer ribonucleic acid methionine complex). Este complexo liga-se à subunidade ribossomal 40S e a outros eIF, incluindo o eIF4E, que reconhece a estrutura cap. A subunidade menor do ribossoma é, então, posicionada na extremidade 5’ do mRNA para que possa fazer o rastreamento de toda a região não codificante da extremidade 5’ (5’ UTR, do inglês, 5’ -end untranslated region) até que um codão de iniciação em contexto favorável seja identificado. Aí iniciar-se-á a fase de alongamento da tradução com a síntese da cadeia peptídica, após o recrutamento e adição da subunidade ribossomal 60S ao complexo de tradução já existente. Quando o último codão é identificado, a tradução termina e a cadeia peptídica liberta-se do ribossoma. A par com isto, toda a maquinaria de tradução é reciclada, para assegurar novos ciclos de tradução proteica. Em processos biológicos de elevado consumo energético, como a mitose e a diferenciação celular, e em condições de stresse, como a escassez de nutrientes e a hipóxia, esta tradução canónica de proteínas encontra-se comprometida. A redução global da síntese proteica é conseguida, normalmente, pela fosforilação da subunidade α do eIF2 ou pela sequestração do eIF4E, após um dado estímulo, com a consequente inibição da iniciação canónica da tradução. Neste sentido, as reservas energéticas da célula vão ser utilizadas apenas no processo biológico em questão ou na resolução de um dado stresse celular. Tal é conseguido pela expressão seletiva de determinados mRNAs. Estes transcritos, normalmente associados a funções de manutenção da homeostasia celular e a processos-chave da célula, conseguem ser traduzidos para proteína através de mecanismos alternativos de iniciação da tradução. Um desses mecanismos envolve locais de entrada internos do ribossoma (IRES, do inglês internal ribosome entry sites), onde, através de estruturas secundárias no mRNA, de alguns fatores canónicos da tradução e de outras proteínas auxiliares denominadas ITAF (do inglês IRES trans-acting factor), o ribossoma é recrutado para as imediações do codão de iniciação, sem o envolvimento da estrutura cap e sem ser necessário o rastreamento da 5’ UTR do mRNA. Apesar das vantagens deste mecanismo alternativo da iniciação da tradução, uma vez que permite a recuperação da homeostasia celular através da gestão dos recursos energéticos da célula, a tradução mediada por IRES pode também ser nefasta. Por exemplo, as células tumorais aproveitam-se deste mecanismo para ultrapassar as condições adversas que se criam no microambiente tumoral (por exemplo, hipóxia, falta de nutrientes e stresse oxidativo) e proliferar. De facto, muitas das proteínas com expressão desregulada em cancro apresentam IRES no respetivo mRNA. Uma delas é o supressor tumoral p53. A presença de três promotores no gene TP53 leva à expressão de três transcritos: dois longos, que permitem a expressão das isoformas FL-p53 (do inglês full-length p53), Δ40p53 e Δ160p53; e um mais curto, que permite a expressão das isoformas Δ133p53 e Δ160p53. Apesar da existência de dois IRES capazes de regular a expressão das isoformas FL-p53 e Δ40p53 já ser conhecida, tendo já sido caracterizadas e definidas as suas estruturas secundárias, só recentemente foi identificado um IRES capaz de mediar a tradução não canónica do Δ160p53, uma isoforma que aparenta desempenhar funções pró-oncogénicas, na presença de mutações missense no gene TP53. Considerando o exposto, esta tese teve como principal objetivo o estudo da regulação da expressão de isoformas alternativas de proteínas envolvidas no cancro, com especial foco na regulação da expressão da isoforma Δ160p53 através do mais recentemente descrito IRES. Na avaliação da capacidade de uma determinada sequência mediar tradução por IRES é frequente recorrer-se a constructos que contêm dois genes repórteres, designados bicistrónicos. O sistema bicistrónico usado nesta tese contem o gene da luciferase da medusa Renilla reniformis (RLuc, do inglês Renilla luciferase) e o gene da luciferase do pirilampo Photynus pyralis (FLuc, do inglês firefly luciferase). A sequência codificante da RLuc é o primeiro cistrão, sendo por isso traduzida de forma dependente da estrutura cap (tradução canónica), ao passo que a sequência codificante da FLuc forma o segundo cistrão. Entre os dois existe uma estrutura secundária designada hairpin (estrutura em grampo). Deste modo, só ocorrerá tradução da FLuc se a sequência clonada a montante desta conseguir recrutar o ribossoma de forma independente da estrutura cap. O IRES capaz de mediar a tradução não canónica do Δ160p53 localiza-se nos primeiros 432 nucleótidos da região codificante desta isoforma e é inibido pela sua 5’ UTR, ou seja, pela sequência codificante da isoforma Δ133p53 localizada a montante do codão de iniciação do Δ160p53. Tendo isto em mente, propusemo-nos avaliar o efeito das mutações missense mais comuns do TP53 na capacidade de indução da atividade do IRES do Δ160p53 na presença da sua 5’ UTR. Recorrendo, então, ao sistema bicistrónico já descrito, foi possível observar o efeito inibidor da 5’ UTR do Δ160p53 na indução do IRES na linha celular cancerígena HeLa, uma vez que foi observado um aumento significativo da atividade da FLuc na ausência da 5’ UTR do Δ160p53, quando comparado com a atividade desta na presença da 5’ UTR. Mais ainda, de todas as mutações testadas (R175H, R248Q, R273H e R282W), apenas a mutação R175H conseguiu reverter de forma significativa parte do efeito inibitório da 5’ UTR, em condições de stresse induzidas pela Thapsigargina, uma droga que induz stresse do retículo endoplasmático, com consequente fosforilação da subunidade α do eIF2 e inibição da tradução canónica. Tais resultados parecem indicar que as funções oncogénicas da mutação R175H vão além da alteração ou perda de função proteica, uma vez que esta parece também atuar ao nível do mRNA, induzindo a expressão do Δ160p53, uma isoforma que já mostrou ter importância na sobrevivência, proliferação e invasão de células cancerígenas. Continuando a caracterização da regulação da expressão do Δ160p53 pelo seu IRES, pretendíamos ainda identificar proteínas auxiliares da tradução alternativa desta isoforma, recorrendo a um sistema que toma partido das interações entre o RNA e a proteína da cápside do bacteriófago MS2. Clonando sequências de interesse do p53 a montante de repetições da sequência do MS2 e realizando a co-transfeção dessas construções com uma outra que contém a sequência que codifica para a proteína da cápside, seguido de co-imunoprecipitação da cápside e dos mRNAs com as repetições do MS2, seria possível fazer a identificação de novas proteínas reguladoras da expressão das sequências clonadas através de espectrometria de massa. Nesta tese descrevemos várias estratégias de clonagem que foram desenvolvidas para clonar, ainda sem sucesso, as sequências de interesse do p53 a montante de repetições da sequência do MS2, assim como possíveis soluções. Procedemos também a otimizações das condições de imunoprecipitação do Hdm2 (do inglês murine double minute 2 human homolog), uma proteína que interage com alguns mRNAs cuja expressão se encontra frequentemente alterada no cancro, como o XIAP (do inglês X-linked inhibitor of apoptosis protein) e o p53, regulando a sua tradução não canónica. Com esta técnica pretendemos, no futuro, identificar novos mRNAs regulados pelo Hdm2 através da sequenciação daqueles que co-imunoprecipitarem com esta proteína. Neste sentido visamos identificar possíveis novos mRNAs detentores de IRES que também possam ter um papel preponderante no desenvolvimento tumoral. Concluindo todas estas linhas de investigação, esperamos desvendar novos conhecimentos sobre a tradução mediada por IRES, assim como parte do papel deste mecanismo na carcinogénese. Provando-se a importância dos IRES no cancro, novas terapias direcionadas para estas estruturas secundárias do mRNA ou para as proteínas que auxiliam este mecanismo de tradução podem ser desenvolvidas.

Unidades organizacionais

Descrição

Palavras-chave

Contribuidores

Financiadores

Entidade financiadora

Fundação para a Ciência e a Tecnologia

Programa de financiamento

5876

Número da atribuição

UID/Multi/04046/2013

ID