A5018411129- Bioinformatics and Genomic Networks
- Gene expression and cancer classification
- Congenital heart defects research
- Machine Learning in Bioinformatics
- Biomedical Text Mining and Ontologies
- RNA Research and Splicing
- Genomic variations and chromosomal abnormalities
- Genomics and Rare Diseases
- Genomics and Phylogenetic Studies
- RNA modifications and cancer
- Face and Expression Recognition
- Neural Networks and Applications
- RNA and protein synthesis mechanisms
- Advanced Clustering Algorithms Research
- Congenital Heart Disease Studies
- Computational Drug Discovery Methods
- Genomics and Chromatin Dynamics
- Neuroblastoma Research and Treatments
- Genetics, Bioinformatics, and Biomedical Research
- Nuclear Receptors and Signaling
- Epigenetics and DNA Methylation
- Cancer, Hypoxia, and Metabolism
- melanin and skin pigmentation
- Single-cell and spatial transcriptomics
- Telomeres, Telomerase, and Senescence
University of Luxembourg
2019-2023
Luxembourg Institute of Health
2017-2019
Centre National de la Recherche Scientifique
2013-2017
Inserm
2013-2017
Université Claude Bernard Lyon 1
2013-2017
École Normale Supérieure de Lyon
2013-2017
Laboratoire de Biologie et Modélisation de la Cellule
2017
Centre de Recherche en Cancérologie de Lyon
2013-2016
KU Leuven
2006-2015
iMinds
2015
A major bottleneck in our understanding of the molecular underpinnings life is assignment function to proteins. While experiments provide most reliable annotation proteins, their relatively low throughput and restricted purview have led an increasing role for computational prediction. However, assessing methods protein prediction tracking progress field remain challenging.We conducted second critical assessment functional (CAFA), a timed challenge assess that automatically assign function....
The RNA helicases DDX5 and DDX17 are members of a large family highly conserved proteins that involved in gene-expression regulation; however, their vivo targets activities biological processes such as cell differentiation, which requires reprogramming programs at multiple levels, not well characterized.Here, we uncovered mechanism by cooperate with heterogeneous nuclear ribonucleoprotein (hnRNP) H/F splicing factors to define epithelial-and myoblast-specific subprograms.We then observed...
This paper presents a novel optimized kernel k-means algorithm (OKKC) to combine multiple data sources for clustering analysis. The uses an alternating minimization framework optimize the cluster membership and coefficients as nonconvex problem. In proposed algorithm, problem are all based on same Rayleigh quotient objective; therefore converges locally. OKKC has simpler procedure lower complexity than other algorithms in literature. Simulated real-life fusion applications experimentally...
E ndeavour ( http://www.esat.kuleuven.be/endeavourweb ; this web site is free and open to all users there no login requirement) a resource for the prioritization of candidate genes. Using training set genes known be involved in biological process interest, our approach consists (i) inferring several models (based on various genomic data sources), (ii) applying each model rank those candidates against profile (iii) merging rankings into global ranking In present article, we describe latest...
Background Housekeeping genes are needed in every tissue as their expression is required for survival, integrity or duplication of cell. commonly have been used reference to normalize gene data, the underlying assumption being that they expressed cell type at approximately same level. Often, terms “reference genes” and “housekeeping interchangeably. In this paper, we would like distinguish between these terms. Consensus growing housekeeping which traditionally data not good genes. Recently,...
Genomic studies and high-throughput experiments often produce large lists of candidate genes among which only a small fraction are truly relevant to the disease, phenotype or biological process interest. Gene prioritization tackles this problem by ranking profiling candidates across multiple genomic data sources integrating heterogeneous information into global ranking. We describe an extended version our gene method, Endeavour, now available for six species 75 sources. The performance (Area...
Abstract Motivation: Hunting disease genes is a problem of primary importance in biomedical research. Biologists usually approach this two steps: first set candidate identified using traditional positional cloning or high-throughput genomics techniques; second, these are further investigated and validated the wet lab, one by one. To speed up discovery limit number costly lab experiments, biologists must test starting with most probable candidates. So far, have relied on literature studies,...
This paper introduces the notion of optimizing different norms in dual problem support vector machines with multiple kernels. The selection yields extensions kernel learning (MKL) such as L∞, L1, and L2 MKL. In particular, MKL is a novel method that leads to non-sparse optimal coefficients, which from sparse coefficients optimized by existing L∞ method. real biomedical applications, may have more advantages over integration for thoroughly combining complementary information heterogeneous...
Gene prioritization aims at identifying the most promising candidate genes among a large pool of candidates-so as to maximize yield and biological relevance further downstream validation experiments functional studies. During past few years, several gene tools have been defined, some them implemented made available through freely web tools. In this study, we aim comparing predictive performance eight publicly on novel data. We performed an analysis in which 42 recently reported disease-gene...
Alternative splicing is the main mechanism of increasing proteome diversity coded by a limited number genes. It well established that different tissues or organs express variants. However, are composed common major cell types, including fibroblasts, epithelial, and endothelial cells. By analyzing large-scale data sets generated The ENCODE Project Consortium after extensive RT-PCR validation, we demonstrate each three types expresses specific program independently its organ origin....
We report that CD47 was upregulated in different EMT-activated human breast cancer cells versus epithelial MCF7 cells. Overexpression of SNAI1 or ZEB1 activated EMT and while siRNA-mediated targeting mesenchymal MDA-MB-231 reversed strongly decreased CD47. Mechanistically, by binding directly to E-boxes the promoter. TCGA METABRIC data sets from patients revealed correlated with Vimentin. At functional level, were less efficiently phagocytosed macrophages vs. The phagocytosis rescued using...
Parkinson's disease (PD) is a heterogeneous disorder, and among the factors which influence symptom profile, biological sex has been reported to play significant role. While males have higher age-adjusted incidence are more frequently affected by muscle rigidity, females present often with disabling tremors. The molecular mechanisms involved in these differences still largely unknown, an improved understanding of relevant may open new avenues for pharmacological modification. To help address...
PINTA (available at http://www.esat.kuleuven.be/pinta/ ; this web site is free and open to all users there no login requirement) a resource for the prioritization of candidate genes based on differential expression their neighborhood in genome-wide protein–protein interaction network. Our strategy meant biological medical researchers aiming identifying novel disease using specific data. supports both gene (starting from user defined set genes) as well available five species (human, mouse,...
Disease-gene identification is a challenging process that has multiple applications within functional genomics and personalized medicine. Typically, this involves both finding genes known to be associated with the disease (through literature search) carrying out preliminary experiments or screens (e.g. linkage association studies, copy number analyses, expression profiling) determine set of promising candidates for experimental validation. This requires extensive time monetary resources. We...
Transcriptomic genome-wide analyses demonstrate massive variation of alternative splicing in many physiological and pathological situations. One major challenge is now to establish the biological contribution physiological- or pathological-associated cellular phenotypes. Toward this end, we developed a computational approach, named “Exon Ontology,” based on terms corresponding well-characterized protein features organized an ontology tree. Exon Ontology conceptually similar Gene...
Estrogen and androgen receptors (ER AR) play key roles in breast prostate cancers, respectively, where they regulate the transcription of large arrays genes. The activities ER AR are controlled by networks protein kinases transcriptional coregulators, including Ddx5 its highly related paralog Ddx17. Ddx17 RNA helicases also splicing regulators. Here, we report that master regulators estrogen- androgen-signaling pathways controlling both upstream downstream receptors. First, required for...
Parkinson’s disease (PD) is a neurodegenerative with unknown cause in the majority of patients, who are therefore considered “idiopathic” (IPD). PD predominantly affects dopaminergic neurons substantia nigra pars compacta (SNpc), yet pathology not limited to this cell type. Advancing age main risk factor for development IPD and greatly influences function microglia, immune cells brain. With increasing age, microglia become dysfunctional release pro-inflammatory factors into extracellular...
Genetic studies (in particular linkage and association studies) identify chromosomal regions involved in a disease or phenotype of interest, but those often contain many candidate genes, only few which can be followed-up for biological validation. Recently, computational methods to (prioritize) the most promising candidates within region have been proposed, they are usually not applicable cases where little is known about (no confirmed fragmentary understanding cascades involved). We seek...
Computational gene prioritization methods are useful to help identify susceptibility genes potentially being involved in genetic disease. Recently, text mining techniques have been applied extract prior knowledge from text-based genomic information sources and this can be used improve the process. However, effect of various vocabularies, representations ranking algorithms on for is still an issue that requires systematic comparative studies. Therefore, a benchmark study about by discussed...