Dissertação (mestrado)—Universidade de Brasília, Instituto de Ciências Exatas, Departamento de Ciência da Computação, 2011. / Submitted by Gabriela Ribeiro (gaby_ribeiro87@hotmail.com) on 2011-06-22T20:32:22Z No. of bitstreams: 1 2011_CarlosViníciusSarmentoSilva.pdf: 3795960 bytes, checksum: 63d12d2765cdb654dc26657b9ec5bfd3 (MD5) / Approved for entry into archive by Guilherme Lourenço Machado(gui.admin@gmail.com) on 2011-06-28T13:01:49Z (GMT) No. of bitstreams: 1 2011_CarlosViníciusSarmentoSilva.pdf: 3795960 bytes, checksum: 63d12d2765cdb654dc26657b9ec5bfd3 (MD5) / Made available in DSpace on 2011-06-28T13:01:49Z (GMT). No. of bitstreams: 1 2011_CarlosViníciusSarmentoSilva.pdf: 3795960 bytes, checksum: 63d12d2765cdb654dc26657b9ec5bfd3 (MD5) / O trabalho de auditoria governamental tem sido realizado no âmbito do Poder Executivo Federal pela Controladoria-Geral da União. Várias estratégias são utilizadas visando a prevenção e o combate à corrupção. No entanto, algumas atividades tais como detecção de cartéis em licitações são limitadas devido à complexidade de se correlacionar informa ções para geração de conhecimento útil para os auditores através da análise de bases de dados. A área de Mineração de Dados tem sido alvo de várias pesquisas tendo bons resultados no processo de descoberta de conhecimento em grandes bases de dados onde várias técnicas já foram de_nidas nesta área tais como classi_cação, clusterização e regras de associação. Sistema Multiagentes por sua vez, apresenta consideráveis vantagens no sentido de possibilitar a distribuição do processamento e fazer uso de autonomia de agentes de softwares para realização de tarefas complexas. Essas duas áreas de estudo, até recentemente separadas, são integradas neste trabalho através de AGent Mining Integration (AGMI), uma arquitetura que integra diferentes técnicas de mineração de dados utilizando uma abordagem multiagentes para automatização do processo de descoberta de conhecimento. AGMI é composta por agentes que operam em três diferentes camadas: estratégica, tática e operacional. Através da autonomia de agentes, AGMI é capaz de integrar técnicas de mineração de dados de forma distribuída e utilizar heurísticas para melhoramento do conhecimento encontrado. Neste trabalho é apresentado um protótipo do AGMI que foi testado com dados reais de licitações extraídas do Sistema ComprasNet. Vários experimentos foram realizados explorando os aspectos de distribuição do processamento e autonomia dos agentes. AGMI apresentou bons resultados quanto ao desempenho, capacidade autônoma de melhorar o conhecimento descoberto e quanto à qualidade do conhecimento apresentado. Comparando com a abordagem testada, utilizando apenas o algoritmo de Regras de Associação, os experimentos com AGMI mostraram um aumento de 170% na qualidade média das 10 melhores regras encontradas e de 350% na qualidade média das 100 melhores regras encontradas. Além disso, AGMI aumentou a qualidade de 193 regras, através de heurística aplicada autonomamente pelo agente Avaliador. As regras descobertas nos experimentos foram analisadas por especialistas da Controladoria- Geral da União e apresentaram fortes indícios de irregularidades em licitações tais como cartéis, simulação de concorrência e direcionamento de editais. _________________________________________________________________________________ ABSTRACT / In Brazil, government auditing is performed by the O_ce of the Comptroller General (CGU), where several approaches are being used to prevent and _ght corruption. However, some activities such as government purchasing fraud detection are limited by the di_culty in _nding e_ective ways to implement. The main problem focused by this research project is how to extract and generate useful knowledge from huge databases of Brazilian Federal procurement processes, in order to help the governmental auditing work. However, activities like detection of cartels are a complex problem in many senses. In terms of _nding useful auditing knowledge, because of the volume of data to correlate information, and also because of the dynamism and diversi_ed strategies used by companies to hide their fraudulent operations. In this research, we combine two originally separated areas and increasingly interrelated: distributed multi-agent systems and data mining. In our approach, we prove the interaction features in a bilateral and complementary way, by introducing AGMI - an AGent-MIning tool for automate knowledge discovery process in a distributed way. Considering the data mining perspective, we have used di_erent model functions, such as clusterization and link analysis with association rules. Autonomous agents are also used in the process in order to improve the discovered knowledge quality. To validate the usage of AGMI, we have performed several experiments using real data from ComprasNet, a government purchasing system of Brazil. Our approach resulted in expressive discovered knowledge. Considering a tested approach using only Association Rule algorithm, the AGMI's experiments have shown a rule quality improvement of 170% in the top 10 rules and 350% in the top 100 rules. Besides, AGMI also enhanced the quality of 193 rules through the autonomous heuristics of Evaluator Agent. According to the auditing experts, the discovered knowledge shall help the work of the CGU auditors in the detection, prevention and monitoring of cartels acting in public procurement processes.

Tese (doutorado)—Universidade de Brasília, Faculdade de Tecnologia, Departamento de Engenharia Elétrica, 2013. / Submitted by Alaíde Gonçalves dos Santos (alaide@unb.br) on 2014-01-21T12:46:46Z No. of bitstreams: 1 2013_BrunoWerneckPintoHoelz.pdf: 3440475 bytes, checksum: 3084aa1ec80c1c223b65cf14f4cac8b2 (MD5) / Approved for entry into archive by Guimaraes Jacqueline(jacqueline.guimaraes@bce.unb.br) on 2014-02-25T12:50:23Z (GMT) No. of bitstreams: 1 2013_BrunoWerneckPintoHoelz.pdf: 3440475 bytes, checksum: 3084aa1ec80c1c223b65cf14f4cac8b2 (MD5) / Made available in DSpace on 2014-02-25T12:50:23Z (GMT). No. of bitstreams: 1 2013_BrunoWerneckPintoHoelz.pdf: 3440475 bytes, checksum: 3084aa1ec80c1c223b65cf14f4cac8b2 (MD5) / Modelos computacionais de confiança e reputação são elementos-chave no projeto de sistemas multiagente abertos. Eles oferecem um meio de avaliar e reduzir o risco de cooperação na presença de incerteza. No entanto, os modelos propostos na literatura não consideram os custos envolvidos na sua aplicação e como os modelos são afetados pela dinamicidade do ambiente. Neste trabalho, um metamodelo para adaptação de confiança a e reputação em sistemas multiagente dinâmicos é proposto. O metamodelo tem como finalidade complementar os modelos de confiança e reputação já existentes, permitindo que agentes deliberativos possam raciocinar sobre os componentes do modelo em uso e reagir a mudança as no ambiente. O processo de adaptação é realizado ajustando a configuração do modelo adotado para melhor se adequar às condições atuais. É demonstrado como o metamodelo pode ser aplicado a modelos propostos na literatura e como planos de adaptação podem ser utilizados para ajustar seus componentes dinamicamente para melhorar seu desempenho. Um mecanismo de aprendizagem, incluindo uma prova de conceito baseada em algoritmos genéticos, é proposto para identificar novos planos de adaptação para cenários similares. Por fim, a avaliação experimental da aplicação do metamodelo e do mecanismo de aprendizagem mostra melhorias significativas em comparação com o uso de modelos não adaptáveis, o que contribui para a melhoria do projeto de agentes autônomos para sistemas multiagente dinâmicos. _______________________________________________________________________________________ ABSTRACT / Computational trust and reputation models are key elements in the design of open multi-agent systems. They offer a way of evaluating and reducing risks of cooperation in the presence of uncertainty. However, the models proposed in the literature do not consider the costs they introduce and how they are affected by dynamic environments. In this work, a meta-model for trust and reputation adaptation in dynamic multi-agent systems is proposed. The meta-model acts as a complement to trust and reputation models, by allowing deliberative agents to reason about the components of the model being used, and to react to changes in the environment. The adaptation process is made by adjusting the model's configuration to better ft the current conditions. It is demonstrated how the meta-model can be applied to existing models proposed in the literature, and how adaptation plans can be used to adjust its components dynamically to improve its performance. A learning mechanism, along with a proof of concept implementation based on genetic algorithms, is proposed to identify new adaptation plans for similar scenarios. Finally, the experimental evaluation of the meta-model application and its learning mechanism shows significant improvements in comparison to the use of non-adaptable models. This contributes to improving the design of autonomous agents for dynamic multi-agent systems.

Spinal Muscular Atrophy (SMA) is an autosomal, recessive form of childhood motor neuron disease and the most common genetic cause of infant mortality in the western world. SMA displays the characteristic hallmarks of a motor neuron disease, including loss of motor neurons in the spinal cord and atrophy of skeletal muscles. However, mounting evidence suggests that multiple tissues and body systems, beyond the neuromuscular system, are affected in SMA. Previous studies have highlighted alterations in the vascular system in both SMA patients and in a variety of mouse models of the disease, reporting alterations in vessel structure and perfusion abnormalities in peripheral tissues. In this project a detailed morphological investigation of the capillary beds of skeletal muscle and the spinal cord, two of the key pathological tissues in SMA was undertaken. This work was conducted in the Smn-/-;SMN2, Smn-/-;SMN2tg/+ and Smn-/-;SMN2;Δ7 mouse models of SMA. Significant alterations in the form and extent of the skeletal muscle and spinal cord capillary bed in SMA mice were identified, the most striking of which being a reduction in capillary density in SMA tissue when compared to control littermate tissue. In skeletal muscle, this reduction in capillary density was found to be a postnatal phenomenon, which occurred independently of denervation, in a variety of phenotypically distinct muscles and in all three SMA mouse models investigated. In the spinal cord, the capillary defect was seen to develop in a similar postnatal pattern to that observed in skeletal muscle. Importantly, a reduction in capillary density was observed in the ventral horn of the spinal cord, which houses motor neuron cell bodies, a known pathological target in SMA. These motor neurons were seen to be surrounded by fewer capillaries than their control counterparts. Using an injectable marker of hypoxia, it was determined that the cells of the ventral horn of SMA spinal cords are hypoxic. This suggests that the capillary defect identified has a functional impact on the tissues it is observed in. Having established the presence of capillary defect in SMA tissue, the effect of potential SMA therapeutics on the capillary defect was then investigated. The effect of HDAC inhibitors, which have been successfully shown to increase the levels of the disease causing Smn protein, was investigated. Treatment with the HDAC inhibitor SAHA was found to ameliorate the capillary defect, significantly improving capillary density in SMA skeletal muscle. This implies that the capillary defect is related to Smn levels in tissue and is amenable to therapeutics which increase Smn levels. Having characterised the capillary defect in SMA tissues in detail, a selection of tools were then used to investigate the underlying mechanisms resulting in the defect. First, using primary cell cultures, the growth and morphology of the key cellular component of capillaries, the endothelial cell, was examined. While displaying reduced levels of the Smn protein, endothelial cells isolated from SMA tissues showed no difference in growth rate, morphology or endothelial cell marker expression when compared to endothelial cells isolated from control tissue. This suggests that the defects seen in SMA capillary beds are not the result of defects in the structure and growth of endothelial cells. Second, retinas from SMA mice were found to exhibit similar capillary defects to those observed in SMA skeletal muscle and spinal cord. Given the entirely postnatal development of the retinal capillary network, the retina was identified as a useful experimental preparation for the further investigation of the mechanisms underlying the capillary defect in SMA. In summary, this work highlights the incidence and importance of capillary defects in mouse models of spinal muscular atrophy.

The misregulation of post-transcriptional mechanisms has been linked to the development and progression of numerous human diseases, in particular neurological disorders and cancer. Investigating these misregulated RNA pathways is essential to fully understand the disease mechanisms, identify novel biomarkers, and to develop effective therapies. In this thesis, I present three manuscripts that investigate the mechanisms behind the post-transcriptional misregulation of RNA in human disease, with a focus on pre-mRNA splicing. In the first manuscript (Bondy-Chorney et al., 2016a), we investigated the role of Staufen1 (Stau1) in splicing regulation in the neuromuscular disorder Myotonic Dystrophy Type 1 (DM1). Here we report the first insights into the mechanism that Stau1 uses to regulate the alternative splicing of INSR exon 11 through an interaction with Alu elements located in intron 10. Moreover, using a high-throughput RT-PCR screen, we uncovered a number of additional Stau1-regulated alternative splicing events in both wild-type and DM1 myoblast cell lines. As Stau1 is known to be aberrantly upregulated in DM1 skeletal muscle, our findings suggest that Stau1 acts as a disease modifier in this disorder. The second manuscript (Sanchez, Bondy-Chorney et al., 2015), describes a novel role of the protein methyltransferase Coactivator-Associated Methyltransferase-1 (CARM1), a protein found to be overexpressed in Spinal Muscular Atrophy (SMA). We found that CARM1 can act as a mediator in the nonsense-mediated decay pathway (NMD) and associated UPF1 to promoted its occupancy on PTC-containing transcripts. We identified a subset of natural non-PTC containing NMD targets that were dependent on CARM1, a number of which were misregulated in SMA. This work uncovered a novel role for CARM1 in the NMD pathway and revealed that defective targeting of PTC-containing mRNAs should be included in the complex array of molecular defects associated with SMA. Finally, the third manuscript (Bondy-Chorney et al., – in prep) examines the alternative iv splicing regulation of the Protein Arginine Methyltransferase PRMT1 exon 2, an event shown to alter the growth, survival, and invasion of breast cancer cells. Here, we used an RNA interference (RNAi) RT-PCR screen to uncover several splicing proteins that regulate the inclusion of exon 2, several of which we found to be misregulated in a panel of breast cancer cell lines and patient tumours. These findings confirmed that the inclusion of PRMT1 exon 2 was regulated by alternative splicing via splicing factors that are altered in breast cancer. Moreover, depletion of one of these splicing factors, RALY, resulted in a decrease in the motility and invasive potential of an aggressive breast cancer cell line. These three manuscripts represent a collection of work focused on elucidating the mechanisms involved in post-transcriptional misregulation of RNA in three diverse human diseases. Taken together, the data presented here highlight the broad impact that proteins, such as Stau1 and CARM1, can have in neuromuscular disorders. Moreover, we also uncovered novel misregulation of splicing proteins that alter alternative splicing patterns in breast cancer. Elucidating these mechanisms is of the highest importance in order to identify potential new and effective treatment avenues.

No description available.

The autosomal recessive disorder spinal muscular atrophy (SMA) causes motor neuron degeneration and muscle wasting, progressing to paralysis and death in severe cases. The disease is caused by deficiency of survival motor neuron protein (SMN) due to deletion or mutation of the SMN1 gene. We seek to develop a protein-based therapy for SMA using an adenoviral vector which encodes a secretable form of SMN fused to a protein transduction domain (PTD) derived from the trans-acting activator of transcription (TAT) from HIV. We generated secretable GFP proteins using transient transfection in mammalian cells and determined that the secretory peptide was inefficient when paired with the native PTD. We generated TAT-GFP proteins in bacteria and observed that the variant TAT3 most reliably tranduced cells in vitro. We did not observe uptake of the therapeutic protein following infection with an adenoviral vector and subsequent secretion of the protein from infected cells.

This work presents a novel approach to arranging shape memory alloy (SMA) wires into a functional heat engine. Significant contributions include the design itself, a preliminary analytical model and the realization of a research prototype; thereby, laying a foundation from which to base refinements and seek practical applications. Shape memory alloys are metallic materials that, if deformed when cold, can forcefully recover their original, "memorized" shapes, when heated. The proposed engine consists of a set of SMA wires stretched between two crankshafts, synchronized to rotate in the same direction. Cranks on the first crankshaft are slightly longer than cranks on the second. During operation, the engine is positioned between two distinct thermal reservoirs such that half of its wires are heated while the other half are cooled. Wires on the hot side attempt to contract, driving the engine in the direction that relieves the heat-induced stress. Wires on the cold side soften and stretch as the engine rotates. Because the force generated during heated recovery exceeds that required for cooled deformation, the engine is capable of generating shaft power. Limited experimental measurements of shaft speed were performed. An analytical model of the engine predicts that the maximum output power for the prototype, under test conditions, should be 0.75 W. Thermal efficiency, though not measured or calculated in this work, is expected to be low. Potential applications may include the conversion of waste heat into shaft power. / Master of Science

Spinal Muscular Atrophy (SMA) is an autosomal recessive neurodegenerative disease, which, before the approval of therapies, was the leading genetic cause of infant mortality. The primary features of this pathology are progressive muscle weakness and atrophy, due to the degeneration of α-motor neurons in the anterior horn of the spinal cord. SMA is caused by deletions or mutations in the Survival Motor Neuron gene (Smn1), which induce reduced levels of the SMN protein. Since 1999, this disease has been primarily associated with splicing defects caused by loss of SMN protein due to its role in ribonucleoparticle biogenesis. However, further research revealed that this mechanism alone is not sufficient to explain the pathogenesis of the disease. More recent findings revealed that deficient SMN levels lead to defective translation in primary motor and cortical neurons, and in multiple tissues at the late stage of disease in the severe Taiwanese mouse model of SMA. Furthermore, SMN protein has been confirmed to be a ribosome-associated protein in vitro, in mouse cell lines and in vivo, and to physiologically regulate the translation of a particular subset of transcripts (defined as SMN-specific transcripts), which are characterized by specific sequence features. Upon SMN loss, the translation of this subset of transcripts is defective. SMN protein is ubiquitously expressed and its levels vary at different developmental stages and tissues in physiological conditions, leading to the hypothesis that translational defects may vary accordingly. However, the effect of SMN loss on translation across different tissue types, SMA mouse models, and disease stages is yet to be clarified. To investigate the link between SMN loss and translational defects in SMA, I took advantage of ribosome profiling to obtain the translatome from multiple tissues, stages and disease mouse models. Given that SMN is ubiquitously expressed, brain, spinal cord and liver were collected to investigate if common features underly translational defects upon its loss in these tissues. Since little is known about how translational impairments are modulated over time, tissues were collected from various developmental and disease stages, ranging from the embryo to the post-natal early-symptomatic stage of SMA. Furthermore, translation defects were studied in multiple models of SMA have ranging from severe to mild (i.e., Taiwanese, Delta7 and Smn2b/-), allowing for the exploration of the heterogeneity of the SMA clinical phenotype. Hence, the tissues were collected from three SMA mouse models (i.e., Taiwanese, Delta7, and Smn2b/-), allowing for the investigation of translational impairments in conditions that range from severe to mild SMA. A wide range of computational approaches was adopted to analyze ribosome profiling data from multiple perspectives, including Principal Component Analysis (PCA), pipelines for the analysis of RiboSeq positional information, differential and Gene Ontology enrichment analysis, and network methodologies. This set of tools applies to the study of ribosome profiling data and allows to investigate the translational mechanisms underlying SMA. This multilevel analysis holds difficulties in the representation and interpretation of the obtained results due to the number of variables (i.e., tissue, stage, model, and disease condition). I hence developed an R package to support the visualization of changes occurring in omics data from complex experimental designs. Next, I focused on the identification of translational defects in SMA through pairwise differential analyses performed on each set of experiments. This allowed me to identify significantly altered transcripts within each comparison. Despite poor overlaps between the sets of translationally dysregulated transcripts across the different stages, tissues, and models, commonly enriched biological processes were found. The analysis of sequence features on translationally dysregulated transcripts across all the stages, tissues, and models revealed the presence of features similar to those already found on the SMN-specific transcripts. In addition, based on network methodologies, I investigated the system-wide effects of SMN loss on connectivity patterns at the translational level, by taking advantage of network-based methodologies to integrate all sets of experiments and unravel any relationships between genes at the translatome level. Causal-inference networks, coupled with differential network analysis, complemented the standard differential analysis by modeling how the fluctuations in reciprocal transcript-specific ribosome occupancy might influence each other. This allowed to detect disrupted relationships in the disease condition across the multiple tissues, stages and models. In summary, this thesis provides, to my knowledge, the first multi-tissue, -stage, and -model translatome analysis to investigate the mechanisms underlying SMA. Furthermore, results provided within this work confirm that translation dysregulation is a common feature of SMA pathology across multiple tissues, stages, and SMA models. This highlights that the presence of specific sequence features of translationally dysregulated transcripts is a common link between defective translational regulation and SMN loss. Moreover, the detection of disrupted connectivity patterns at the translatome level underlies that a strong remodeling occurs upon SMN loss, and further emphasizes the pivotal role of this protein in translation. These outcomes highlight the importance of further investigating the mechanisms underlying defective translation in SMA from a system perspective to provide a comprehensive understanding of this pathology and promote the development of effective therapeutic strategies.

As misturas asfálticas do tipo SMA apresentam granulometria descontínua, composta por uma maior fração de agregados graúdos, uma rica massa de ligante/fíler (mastique) e aproximadamente 4% de volume de vazios. Possuem um esqueleto pétreo de alta estabilidade devido ao contato pedra-pedra, que proporciona uma maior resistência à deformação permanente. Geralmente o teor de ligante asfáltico é superior a 6%, formando uma película asfáltica mais espessa. São utilizadas fibras para evitar o escorrimento do ligante durante as etapas de produção e lançamento e, geralmente, são usados asfaltos modificados por polímero. Esta pesquisa tem por objetivo avaliar se o ligante asfalto-borracha possibilita misturas asfálticas SMA capazes de atender aos valores limites de aceitação e, por meio de ensaios de laboratório (resistência à tração, módulo de resiliência, fadiga e deformação permanente em simulador de tráfego), comparar o comportamento de misturas SMA com diferentes tipos de ligante (asfalto convencional CAP 20, modificado por polímero e asfalto-borracha) com um concreto asfáltico convencional de granulometria contínua (Faixa C do DNER). Os resultados obtidos apresentam o comportamento de uma mistura SMA com asfalto-borracha muito mais próximo de misturas SMA produzidas com um ligante modificado por polímero do que com um asfalto convencional. / SMA is a gap-graded asphalt mixture with a large proportion of high quality coarse aggregate, a high content of mastic (binder/filler), and approximately 4% of air voids. The larger proportion of coarse aggregate provides a greater stone-to-stone contact, which results in a mixture more resistant to permanent deformation than the conventional Hot Mix Asphalt (HMA). The asphalt content is typically greater than 6.0 percent, which increases the film thickness. Fibers are used to prevent drainage of the asphalt binder during the HMA production and placement, and polymer-modified asphalt cements are usually used. This research aims to evaluate if an asphalt-rubber binder produces SMA mixtures able to meet the technical requirements. The behavior of SMA mixtures produced with different binders (conventional AC-20, polymermodified, and asphalt-rubber) is analyzed based on laboratory tests (tensile strength, resilient modulus, fatigue, and permanent deformation in a traffic simulator) and compared to the behavior of a conventional dense-graded HMA The results show that the behavior of SMA mixtures produced with asphalt-rubber is much closer to SMA mixtures produced with polymer-modified binder than conventional asphalts.

As misturas asfálticas do tipo SMA apresentam granulometria descontínua, composta por uma maior fração de agregados graúdos, uma rica massa de ligante/fíler (mastique) e aproximadamente 4% de volume de vazios. Possuem um esqueleto pétreo de alta estabilidade devido ao contato pedra-pedra, que proporciona uma maior resistência à deformação permanente. Geralmente o teor de ligante asfáltico é superior a 6%, formando uma película asfáltica mais espessa. São utilizadas fibras para evitar o escorrimento do ligante durante as etapas de produção e lançamento e, geralmente, são usados asfaltos modificados por polímero. Esta pesquisa tem por objetivo avaliar se o ligante asfalto-borracha possibilita misturas asfálticas SMA capazes de atender aos valores limites de aceitação e, por meio de ensaios de laboratório (resistência à tração, módulo de resiliência, fadiga e deformação permanente em simulador de tráfego), comparar o comportamento de misturas SMA com diferentes tipos de ligante (asfalto convencional CAP 20, modificado por polímero e asfalto-borracha) com um concreto asfáltico convencional de granulometria contínua (Faixa C do DNER). Os resultados obtidos apresentam o comportamento de uma mistura SMA com asfalto-borracha muito mais próximo de misturas SMA produzidas com um ligante modificado por polímero do que com um asfalto convencional. / SMA is a gap-graded asphalt mixture with a large proportion of high quality coarse aggregate, a high content of mastic (binder/filler), and approximately 4% of air voids. The larger proportion of coarse aggregate provides a greater stone-to-stone contact, which results in a mixture more resistant to permanent deformation than the conventional Hot Mix Asphalt (HMA). The asphalt content is typically greater than 6.0 percent, which increases the film thickness. Fibers are used to prevent drainage of the asphalt binder during the HMA production and placement, and polymer-modified asphalt cements are usually used. This research aims to evaluate if an asphalt-rubber binder produces SMA mixtures able to meet the technical requirements. The behavior of SMA mixtures produced with different binders (conventional AC-20, polymermodified, and asphalt-rubber) is analyzed based on laboratory tests (tensile strength, resilient modulus, fatigue, and permanent deformation in a traffic simulator) and compared to the behavior of a conventional dense-graded HMA The results show that the behavior of SMA mixtures produced with asphalt-rubber is much closer to SMA mixtures produced with polymer-modified binder than conventional asphalts.