Luxembourg National Research Fund

The opioid system regulates pain, reward and emotional behaviour through a family of peptides produced in the brain commonly known as “endorphins” that interact with membrane receptors. There are four classical opioid-peptide-binding receptors, which are the main target or the extensively prescribed painkillers morphine, oxycodon and fentanyl. While these drugs are unrivalled in managing acute pain or providing palliative care in cancer, their efficacy is limited by severe side effects such as dependence or respiratory depression, which can lead to overdoses and death and preclude their use in the context of chronic pain treatment. Their increased use and misuse both in and out of the clinic has contributed to the current opioid crisis, and the need to develop new analgesic drugs with different mechanisms of action and reduced complications is more critical than ever.

The team at the Luxembourg Institute of Health identified a novel receptor for opioid peptides, namely ACKR3, with unconventional negative regulatory functions. ACKR3 is termed “scavenger” receptor and acts as a sink for endogenous analgesic or euphoric opioid peptides, hence limiting their availability to the classical opioid receptors. We also developed a modulator of ACKR3, called LIH383 with remarkable properties and were able to show that by blocking the receptor scavenging function, opioid peptide levels can be restored and hence their natural beneficial effects. ACKR3 represents therefore a relevant and valuable alternative target, for the development of safer therapeutic strategies for opioid-related disorders. However, to fully exploit the therapeutic potential of ACKR3 in pain treatment, a better understanding of its mechanisms of action and the design of new drugs recapitulating LIH383 activity towards ACKR3 but with improved pharmacological profile are particularly needed. Moreover, the activity of these newly generated compounds also needs to be validated using reliable in-vivo pain models.

The aims of this project are therefore (1) to investigate the regulatory roles of ACKR3 in the opioid system, (2) to improve the properties of ACKR3 modulators and (3) to validate ACKR3 as a new alternative target for pain management and opioid-related disorders. In particular, expertise in medicinal chemistry and access to state-of-the-art cellular tools and in-vivo pain models are essential to make this project possible and successful. These are not currently available in Luxembourg but are well established in Quebec at the University of Sherbrooke and the University of Montreal. This project relies thus on obvious complementarities between the groups of our consortium.

This project will allow the PhD candidate to (1) conduct her/his own original research and (2) acquire new skills and knowledge for her/his future career, (3) to benefit from the complementary expertise of the involved laboratories and (4) be part of a highly dynamic international research environment. The results of this project will lead to a better understanding of the mechanisms and the relevance of ACKR3 as a regulator of the opioid system. It will advance new concepts such as indirect opioid receptor modulation and lead to the development and validation of next-generation drugs in pain-related models and open new therapeutic avenues for opioid-related disorders. It will also reinforce the interactions, collaborations and scientific exchanges between Luxembourg and Quebec based on true scientific needs and complementarities.

I am text block. Click edit button to change this text. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Ut elit tellus, luctus nec ullamcorper mattis, pulvinar dapibus leo.

The goals of our project are to better understand Zellweger Spectrum Disorders (ZSD) and to develop therapies for patients affected by these inherited diseases. ZSDs are caused by mutations in genes that are needed to form and support the function of compartments in our cells (peroxisomes) that are essential for metabolizing and synthesizing certain lipids. Dysfunctional peroxisomes lead to development of progressive diseases affecting several organs, including the brain and the liver. We do not currently understand precisely how the cellular and metabolic defects lead to organ failure and there is no cure for ZSDs.

With our project we aim to contribute to addressing those knowledge gaps and unmet needs by using a two-fold strategy: 1) mimicking ZSD in the model organism Danio rerio (zebrafish) to understand pathology and 2) identifying drugs that can improve peroxisome functions in patient cell lines, and thus could be trialed in the animal models and ultimately, individuals with ZSD. Zebrafish is an excellent, yet still underutilized, model organism for ZSDs as peroxisome genes are highly conserved between fish and humans. We have successfully manipulated zebrafish via gene editing to disable Pex1, the protein that is most commonly mutated in ZSD patients. This will allow us to analyse down to the molecular detail how the disease develops in a living vertebrate organism. In addition, we used skin cells derived from ZSD patients as a model to test drugs already used for other indications and look for any peroxisomal rescue effect by visualizing a key protein that does not localize properly in patient cells. To generate and analyze the data, we used sophisticated robotics, microscopes, and computer programs to screen hundreds of drugs consistently and efficiently.

This project offers the perspective to advance the overall field by providing a new model to further study the disease and to potentially help patients directly by discovering compounds that act on the disease mechanism.

Handwriting is one of the most powerful expressions of human cognition. We can find many different instantiations of handwriting in everyday movements such as gesturing, pointing, sketching, etc. To advance the field, researchers are becoming interested in computational models that can provide detailed predictions of this kind of movements, taking into account a large number of interrelated factors. With such computational models, researchers can study the performance of efficient, coordinated, goal-directed movement patterns involving multiple body systems and multiple levels within the nervous system.

This AFR-Bilateral project will investigate new techniques to create efficient and robust computational models of handwriting movements using Deep Learning (DL), informed by the Lognormality Principle [Plamondon 2021]: The emergent lognormality of the neuromuscular impulse response of a given human motor system is a basic global feature reflecting the behavior of individuals who are in perfect control of their movements. The production of complex movements is accomplished through the exploitation of this principle, yet there is no efficient implementation of such principle to date. From a general perspective, the project will provide researchers with feature-rich human-like samples which in turn will allow researchers to conduct their own experiments at low risk. Therefore, the project has the potential to impact a large number of domains, since the proposed methodology covers fundamental areas that are transversal to several scientific disciplines.

Because software is pervasive, the user base of software is ever increasing. This results in endless encounters between users and software which more than often end up with some unexpected output or wrong results on the user side. In many cases, users report the bugs to the developers. Oftentimes, however, too many bugs are reported that exceed the human resources and time available for triaging and fixing those bugs. Researchers have therefore proposed techniques to automate several of the bug reports triaging tasks such as prioritizing, classifying, and assigning bug reports. Recently, researchers have started to focus on how to leverage bug reports in actionable software engineering tasks such as fault localization and automatic program repair. Nonetheless, these approaches generally do not actually dig into bug reports: for example, they use token matching between a bug report and source code to perform localization at the file level. We argue that the current utilization of bug reports in software engineering tasks is still suboptimal with great room for improvement and expansion. For example, many bug reports include specific information that would help in generating test cases that could reveal the bug being reported. Such tests are very useful for the purpose of bug localization and fixing.

Our research project BURGEON will focus specifically on test case generation based on bug report information. In particular, we will address two typical problems in automatic testing: finding the right bug triggering inputs, and inferring test oracles.

Surfaces and their deformation (or moduli) spaces play a ubiquitous role in the geometry and topology. This comes from their relationship to different areas of mathematics, but also to the many ways in which they can be viewed. This project aims to study surfaces from multiple directions. Identities, or infinite equations relating lengths of curves and famously used by Mirzakhani to compute volumes of moduli spaces, are the first topic of exploration. Circle patterns, elementary geometric objects with a hidden complexity, are the second object of study, where here the goal is to approach a “complex projective” version of a well-known conjecture. In the final part, surfaces group representations in particular matrix groups (powerful dynamical generalizations of “classical” moduli spaces”) will be studied.

In this project we focus on random graphs, which is a class of mathematical objects playing a fundamental role in modern probability theory, and whose study has an immense scope of applications, ranging from mathematical statistics, to network modeling, mathematical physics and machine learning. There are a lot of challenges which need to be addressed, and their study requires to master a large palette of mathematical tools, stemming e.g. from combinatorics, complex analysis, geometry and stochastic analysis. The main goal of the GRAALS project is to train two outstanding young researchers in order to analyse random graphs through the specific prism of limit theorems and probabilistic approximations. These are two domains of research in which the four PIs involved in the project are world-class leaders, specifically because of their contributions to the so-called Stein’s method for probabilistic approximations.

The project is organized around three work packages: (WP1) Probabilistic approximations and variance estimates, (WP2) Random geometric graphs, and (WP3) Fluctuation theory for graph limits. In the mid-term, it will open up new perspectives and give rise to new challenges for the modelling of real networks. Moreover, the new discoveries will have a scope that goes well beyond the framework of the present project, involving for instance applications in wireless networks, complex networks, statistical mechanics and cosmology.

This bilateral project is natural, because Luxembourg and Singapore represent two research centers that have a role as world leaders in the field of probabilistic approximations and their applications — with skills that complement each other perfectly. It also represents a unique opportunity to consolidate existing links and create new ones, by securing the position of Luxembourg and Singapore at the forefront of the Stein’s method methodology.

Separating hydrogen from natural gas mixtures is an important intermediate technology that allows for the distribution of hydrogen (that could be produced in a variety of different ways, ideally using renewable electricity and electrolysis to produce so-called “green hydrogen”) through the existing natural gas distribution network. While this has many benefits and could be implemented quite rapidly, there are no suitable technologies for the removal of hydrogen from natural gas mixtures at a range of scales – especially at relatively small scales. This project will develop advanced materials with very small pore networks, small enough to separate hydrogen (one of the smallest molecules) from the other molecules present in natural gas, which are usually methane, carbon dioxide, and nitrogen. The hydrogen will be used in the final application, such as industrial heating, chemical manufacturing, or refuelling for heavy vehicles like buses and trucks.

Early robotic missions for exploring the Moon relied on several engineers sitting in the control room around the clock to control the robot remotely. Today, robots are capable to use sensors to get an impression of their environments and drive around autonomously. For this, cameras are used to keep track of the movements of the robot and to help the robot make decisions based on its surroundings. One of the mission goals is to make a map of the Moon. The LUNAR-SLAM project will help the robot to find its own location on the Moon and create a more accurate map. This can be achieved by studying the latest technologies available and combining different sensors such as laser scanners and accelerometer data. This data can be merged with the visual input from the camera to make more accurate estimations of what the environment looks like. Relying on cameras alone can be dangerous as the robot could be blinded by the sunlight or it could be lost when driving in dark areas. When the research phase is finished, a prototype can be developed and tested in the LunaLab of the University of Luxembourg and at a testing facility at NASA Ames in California. These tests will allow measuring how well the system will work on the real Moon.