Projets novateurs réalisés

Characterization of pharmacogenetic variants in a Brazilian admixed cohort of elderly individuals

Differences in our genes can increase or decrease susceptibility to certain diseases and influence response to drugs. Pharmacogenomics is the study of how genetic variation impacts efficacy and toxicity of drugs. To date there are more than 1200 pharmacogenes described in literature, but only a small fraction of them have been implemented into clinical practice to inform prescription decisions. In the case of Brazil and Latin American, population admixture between three major ancestral groups (Native American, European, and sub-Saharan African) plays a key role for pharmacogenomics implementation because extrapolation of data derived from studies with well-defined ethnic groups is clearly not applicable to most of the population. Thus, the aim of this project is to asses common and rare variants in clinically relevant pharmacogenes in an admixed cohort of Brazilian individuals and to identify the origin of these variants by mapping the local ancestry. These findings will provide a high resolution map for studies of pharmacogenomics in admixed population and contribute to the characterization of novel rare variants that could impact on functional variability in pharmacogenes.

Machine-Learning assisted computational search for new Li-Ion

The objective of this project is to apply machine-learning algorithms for the assisted computational search of new energy materials. In particular, cathodes for Li-ion batteries and solid-electrolyte interphase systems will be explored.
Development of new high-performance battery materials is an integral component of overcoming the dependence on fossil fuels and ending the energy and climate crisis. Cobalt, the main component in state-of-the-art Li-ion batteries, has already tripled in price in the past few years, and significant further increases are expected with a shift to electric vehicles and battery grid storage. Novel and cheap materials are urgently needed. Materials discovery in chemistry and material science often relies upon trial and error, and thus, proves challenging since a rational design approach is not present. This project tries to tackle this issue while using an assisted computational search approach to find alternative cathode materials.

DSP-Based Digital Controller for a Bi-Directional DC-DC Converter

The objective of this research is to design and implement a controller based on a Digital Signal

Controller (DSC) for the operation of the bi-directional DC-DC converter developed by Alpha

Technologies Ltd. As a result of this work, the intern must provide the partner organization with the

appropriate firmware to run the DSC and its proper documentation. To achieve this goal the intern will

work for a four-month period to complete the following five stages: system modelling, controller

design, DSC peripherals modelling and controller firmware development, DSC-Converter integral

simulation, and experimental implementation and tests. The development and use of modern

methodologies for implementing digital controllers for power converters will help the partner to

develop capabilities to upgrade other power conversion topologies as part of their family of products.

Kant et l’Académie de Berlin

À ce jour, peu d’études ont été réalisées sur la contribution de l’Académie de Berlin à la pensée d’Emmanuel Kant. Pourtant, cette influence est non seulement véritable, elle détermine grandement l’évolution de certains des thèmes centraux de la philosophie kantienne. Pour ces raisons, une thèse faisant l’analyse des écrits de Kant et des personnages gravitant dans le giron de l’Académie de Berlin permettra d’émettre des hypothèses crédibles sur le rôle joué par cette institution dans la genèse des idées de Kant. Un travail de ce type pourrait certainement représenter une contribution utile aux études kantiennes, dans la mesure où il s’efforcera d’offrir une compréhension des écrits kantiens articulée autour des débats métaphysiques et scientifiques de leur époque. Seul un séjour à Berlin m’offre la possibilité de consulter certains mémoires archivés à la Bibliothèque d’État qui sont étroitement liés à mon sujet de recherche. J’espère notamment y trouver une documentation pertinente permettant d’approfondir nos connaissances sur le débat des forces vives. La supervision de Tobias Rosefeldt promet d’enrichir significativement mes démarches. Elle me permettra de côtoyer d’autres chercheurs et étudiants qui partagent mes intérêts de recherche.

Digital literacy technologies: Applying empirical evidence to optimize reading outcomes

Learning to read is the single most important skill that children learn in elementary school. Squiggle Park is an innovative company that designs digital literacy games that allow children to have fun while learning this critical skill. The Language and Literacy Lab at Dalhousie University has international-level scientific expertise in the process of children’s reading development. This MITACS cluster is a partnership between these business and scientific leaders—with the goal of improving the effectiveness of its products. Student interns will develop and validate new content, with an eye towards increasing Squiggle Park’s market share.

Establishing the role of the hippocampus in working memory impairments in children with focal epilepsy

Working memory impairments are a common co-morbidity of pediatric epilepsy and adversely impact school performance and quality of life. Research in adults with temporal lobe epilepsy suggests that brain networks established in long-term memory (i.e., hippocampus in the mesial temporal lobe) may be involved in working memory, but it is not yet known whether these findings extend to children with focal epilepsies. This research will examine the neuroanatomical underpinnings of working memory impairments in children with focal epilepsy. We will use task-based fMRI to examine hippocampal connectivity, and correlate results with cognitive performance and parent-report of working memory. Children with focal epilepsy will exhibit impairments in working memory and greater working memory problems in daily life, compared to typically developing control children, which in turn will be associated with alterations in hippocampal connectivity. This study will contribute to our understanding of memory impairment in pediatric epilepsy, inform pre-surgical evaluation of memory in clinical practice, as well as the development of interventions.

Information extraction from cross lingual short text

The main goal of this proposed project is to develop a novel bilingual topic model, which explicitly models the word co-occurrence cross-lingual in document-aligned comparable data using a novel merging and shuffling strategy, called CL-BTM. Given a document-aligned multilingual corpus, CL-BTM can be employed to extract latent cross-lingual topics that optimally describe the observed data and discover language-specific per-topic word distributions in each language. A novel bilingual topic model is used to obtain the shared global topic distributions and language-specific topic-word distributions. Ideally, the hierarchical representations of text would be well applied for text understanding and classifications. For further application, the topic coherence and the correlation between entities can be accurately extracted in a document using both the local information (represented as biterm) and the global knowledge (topic knowledge) in a knowledge base, by jointly modeling and exploiting the context compatibility.

WP 3.2.2 – Analyse automatisée des journaux

Ciena est une entreprise canadienne de chef de file dans l’ingénierie et la fabrication de systèmes et dispositifs réseaux. L’entreprise compte environ 5 000 produits opérationnels dans son portefeuille. La grande majorité des produits Ciena génèrent des journaux lors du démarrage et des opérations en mode mission à partir des différentes tâches fonctionnant sur leurs systèmes d’exploitation en temps réel. L’entreprise souhaite donc augmenter les capacités de son logiciel afin de pouvoir collecter tout type de données de journal générées sur le site de production et liées à d’autres informations externes afin d’extraire des connaissances exploitables. Dans ce contexte, l’objectif général proposé du projet « Analyse automatisée des journaux » est de fournir à Ciena des capacités avancées d’analyse des journaux, des outils d’exploration de données et des algorithmes d’apprentissage automatique pour soutenir ses opérations de fabrication de produits lors de l’introduction et de la phase de production de nouveaux produits, dans le cadre du cycle de vie du produit ainsi que pour l’analyse des échecs de retour sur le terrain.

Answering questions according a document by using machine reading comprehension techniques.

This proposed project aims to enable computers to read and understand a document and answer questions related to the documents by using Machine Reading Comprehension technology. In other words, the goal of the project is to equip computers with a similar reading comprehension ability as human beings. The technology can be used in many contexts, such as intelligent customer services, chat robots, and intelligent design decision making. The first step of this proposed project will be focused on the implementation and testing of a few popular machine reading comprehension models using existing open data sets. Afterwards, the developed models will be integrated with the question asking algorithms developed by Dr. Zeng’s design lab to support the Environment-Based Design methodology. The outcome of this project will become an integral part of the next generation CAD/E systems.

A ZooMS-informed archaeozoological and taphonomic analysis comparing Neanderthal and Homo sapiens subsistence behaviors in Northwest Italy

This project aims to contribute to the ongoing discussions surrounding the topic of the Middle to Upper Paleolithic Transition in Europe which is marked by two turning points in the evolution of our own species, anatomically modern Homo sapiens (AMHS). While Neanderthals, our closest evolutionary cousins, went extinct, modern humans migrated out of Africa and dispersed very rapidly across Eurasia. While both populations were exposed to the same dramatic climatic shifts at the time, it is only the Neanderthals that quickly disappeared from the archeological records, suggesting that AMHS populations may have been better adapted to react to environmental changes than Neanderthals. Since diet is a good monitor of adaptation, this research compares the subsistence strategies of both human groups as they occupied, in quick succession, the site of Riparo Bombrini (Balzi Rossi, Liguria, Italy). I conducted archeozoological and taphonomic analyses on faunal collections to give the first direct comparison between Neanderthal and AMHS diets and hunting strategies on one of the only site in Northwest Italy entirely documented with modern archeological methods. TO BE CON’T

Sol-gel encapsulation of safflower derived oilbodies and for the evaluation of stability under simulated gastrointestinal conditions and manufacturing processes.

SemBioSys Genetics, Inc is active in the production of recombinant, pharmacologically active proteins using the seeds of the safflower plant. The purified product is associated with oilbodies derived from the seed. In this project, the intern will perform research on a novel encapsulation technology for functional food/feed applications. The technology is aimed at encapsulating a pharmaceutically active product upon the surface of oilbodies in a silica-based hull. This hull will protect the product from stomach acids while releasing the drug in the lower gastrointestinal tract. Using this silica technology will allow for a targeted release, utilizing the enhanced degradation of the silica matrix at high pH levels. This project has the potential to create a new product line for SemBioSys. Additional, information required to file patents for other applications will also be pursued to protect this technology for commercialization.

Optimisation de la puissance et de l’énergie spécifiques des cellules au lithium-ion par métaheuristique combinée à la modélisation

Le projet actuel vise l’amélioration des performances des cellules au lithium-ion servant au stockage d’énergie électrique. L’énergie et la puissance des cellules au lithium-ion font parties de leurs propriétés les plus importantes. Une optimisation sera donc effectuée afin de maximiser l’énergie ainsi que la puissance de ces dispositifs. Le travail sera réalisé pour les trois formats de cellules les plus communs : les formats cylindrique, prismatique et sachet. De plus, le projet étudiera les deux matériaux, ou composés chimiques les plus utilisés actuellement pour le stockage d’énergie (NCA et NMC). Les résultats de l’étude conduiront à diverses configurations de cellules optimisées offrant des performances maximales. Ces cellules pourront être utilisées pour équiper des appareils électroniques ou des véhicules électriques, entre autres. Les configurations obtenues seront donc comparées afin de déterminer lesquelles répondent le mieux à chaque catégorie d’application.