Search

Rationale and objectives: Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer. Despite a 90% five-year survival rate, survivors of childhood ALL often suffer from late effects, including cardiometabolic disorders. Contributing factors such as inflammation and oxidative stress, combined with drug treatments, can induce premature aging and cellular senescence with a significant impact on cardiometabolic disorders. Premature aging can lead to decreased thymic T-cell production, resulting in decreased circulation of T-cell receptor excision circles (TRECs). Because diet has been associated with cardiometabolic disorders, we hypothesized that the quality of diet in children who had survived ALL was related to the immune aging biomarker TREC, in concert with inflammatory status. Methods: Adolescent and young adult survivors of pediatric ALL of the PETALE cohort (n=241, 22.1 ± 6.3 years at diagnosis, 49.4% male) were examined in their profile for TREC levels (by qPCR) and for adherence to 6 diet quality indices. Results: Adjusted linear regressions revealed that the Healthy Diet Indicator (HDI) was associated with TREC levels (β=50.0, p=0.005, adjusted p=0.03). After performing a conceptual relational analysis (CAR) for data mining of various biomarkers of inflammation, oxidative stress, endotoxemia, and endothelial or adipose dysfunction; interleukin-6 (IL-6) and C-reactive protein (CRP) were found to be negatively associated with TREC levels (β= -80 and -80.1, p=0.017 and 0.026, respectively) but not with HDI. Further analysis revealed that IL-6 and CRP levels were moderators, but not mediators, of the association between HDI and TREC. Conclusion: This study supports the positive impact of a healthy diet on premature immune aging and the moderating role of inflammation in this association. Citation Format: Abderrahim Benmoussa, Tibila Kientega, Sophia Morel, Guillaume Cardin, Sophie Bérard, Mickael wajnberg, Petko Valtchev, Alexandre Blondin-Masse, Daniel Curnier, Maja Krajinovic, Caroline Laverdière, Daniel Sinnett, Emile Levy, Sophie Marcoux, Francis Rodier, Valérie Marcil. Poor diet quality is associated with immune aging in survivors of pediatric acute lymphoblastic leukemia [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 2003.

To gain an in-depth understanding of human diseases, biologists typically mine patient data for relevant patterns. Clinical datasets are often unlabeled and involve features, a.k.a. markers, split into classes w.r.t. biological functions, whereby target patterns might well mix both levels. As such heterogeneous patterns are beyond the reach of current analytical tools, dedicated miners, e.g. for association rules, need to be devised. Contemporary multi-relational (MR) association miners, while capable of mixing object types, are rather limited in rule shape (atomic conclusions) while ignoring feature composition. Our own approach builds upon a MR-extension of concept analysis further enhanced with flexible propositionnalisation operators and dedicated MR modeling of patient data. The resulting MR association miner was validated on a pediatric oncology dataset.

International audience; A complex system, integrates multiple sub-systems. (Carney, Fisher, & Place, 2005) Each sub-system contains some form of domain knowledge but the main difference that can be seen between a complex system and a set of simple system is that the knowledge the complex system presents is larger than the sum of the knowledge that each sub-system contains (Billaud, Daclin, & Chapurlat, 2015). In both cases, each sub-system is made of elements that are exploitable together, but two different sub-systems have information that can’t be used in concert. Therefore, the tacit knowledge, the knowledge contained in the interaction between the sub-system, is usually harder to extract than the knowledge contained in the sub-systems (Yahia, Lezoche, Aubry, & Panetto, 2011).In “simple cases”, the sub-system can be presented as a formal context, a cross-table of objects and their attributes, and the links between the diverse sub-systems can be represented as relational context, a cross-table containing the objects of two sub-systems, and modeling if two objects are in relation. In this paper, we aim to present the process to extract knowledge from such a model of complex system. We will show how to apply it to a use case of real data from the neurology domain.

International audience; To make knowledge-supported decisions, industrial actors often need to examine available data for suggestive patterns. As industrial data are typically unlabeled and involve multiple object types, unsupervised multi-relational (MR) data mining methods are particularly suitable for the task. Current MR association miners merely produce singleton-conclusions rules hence might miss multi-way dependencies. Our novel MR miner builds upon a relational extension of concept analysis to extract general associations. While successfully dealing with circularity in data, it avoids producing cyclic rules by limiting the description depth of relational concepts. Our rules’ relevance was validated by an application to aluminum die casting.; Pour prendre des décisions fondées sur la connaissance, les acteurs industriels ont souvent besoin d'examiner les données disponibles pour trouver des modèles suggestifs. Comme les données industrielles sont généralement non étiquetées et impliquent de multiples types d'objets, les méthodes d'exploration de données multi-relationnelles (MR) non supervisées sont particulièrement adaptées à cette tâche. Les chercheurs d'association actuels ne font que produire des règles de conclusions simples et peuvent donc manquer les dépendances à plusieurs voies. Notre nouveau chercheur MR s'appuie sur une extension relationnelle de l'analyse conceptuelle pour extraire des associations générales. Tout en traitant avec succès la circularité des données, il évite de produire des règles cycliques en limitant la profondeur de description des concepts relationnels. La pertinence de nos règles a été validée par une application au moulage sous pression de l'aluminium.

International audience; Interoperability is a major stake for industry, and in general for all the systems, of any dimension, that need to share contents in every shape. It provides that the exchanges between different parts of different entities perform in a perfect way. Various problems could arise and let the interoperation difficult or impossible. One of those problems could be the presence of implicit knowledge in the systems models. This kind of problems can be faced through knowledge formalisation strategies. The Formal Concept Analysis (FCA) is a mathematical tool to represent the information in a structured and complete way. In this scientific work, we present an extension of the FCA, the Relational Concept Analysis, to reveal tacit knowledge hidden in multi contexts systems.

Today, software architecture has a vital role in achieving quality goals for large scale distributed software systems which is made up of components and connectors. Especially, software development mainly consists of composing re-usable components. The main issue in this approach resides in the difficulty to make these heterogeneous components communicate with each other smoothly, especially in mobile and pervasive computing area. These components can be replaced, dropped or added easily using injection of control mechanism (IoC). In this work, we propose and validate a universal connector framework that allows, through a web service interface using cloud technologies, diverse software components and mobile devices communicate with each other. For example, components and devices in pervasive systems can exchange context information, and analyze them to deduce new rules. As a proof of concept, we apply this connector in a mobile messenger context and shows our results.