GBGaladriel Brière

Galadriel Brière

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Galadriel Brière has a total of 42 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Benchmarking the impact of data leakage on the performance of knowledge graph embedding models for biomedical link prediction

    by Galadriel Brière, Thomas Stosskopf, Benjamin Loire, Anaïs Baudot

    Motivation: Knowledge graphs (KGs) organize complex biomedical knowledge into structured representations of entities and relations. Knowledge graph embedding (KGE) models learn compact representations of KGs, and are widely applied for biomedical link prediction. Despite extensive work on KGE models, current evaluations often overlook the issue of data leakage, which can artificially inflate performance and undermine benchmark validity. Data leakage can arise when (i) there is redundancy between training and test sets, (ii) the model leverages illegitimate features, or (iii) the test set does not accurately reflect real-world inference. Results: We assess the impact of data leakage on KGE-based link prediction across three biomedical KGs, using decoder-only and GNN-based models. We first demonstrate the impact of train-test redundancies and implement a systematic procedure to detect and remove them. Using permutation experiments, we also investigate whether node degree acts as an illegitimate predictive feature, and find no evidence that predictions are driven by degree alone. Finally, we evaluate how well common test set sampling strategies reflect real-world inference in drug repurposing. We compare random and cold-start data splits with an independent test set from Orphanet, and observe a substantial performance drop on the latter, indicating that current benchmarking practices may overestimate how well KGE models generalize to practical applications. Overall, our findings highlight the importance of rigorous benchmark design and careful evaluation of the generalization ability of KGE models for biomedical link prediction. Availability and implementation: Code and results are openly available on GitHub at https://github.com/galadrielbriere/data_leakage_kge_benchmark.git and archived on Zenodo at https://doi.org/10.5281/zenodo.21885112.


  • Open Access

    PhenoXtract: combining Large Language Model and Knowledge Graph embedding to extract phenotypes from clinical descriptions

    by Silvia Berardelli, Galadriel Briere, Benjamin Loire, Federica De Paoli, Andrea Maria Gazzo, Ivan Limongelli, Paolo Magni, Susanna Zucca, Anais Baudot

    Motivation: Standardized phenotypic descriptions are essential for accurate diagnosis, yet clinicians and researchers face challenges in manually extracting and mapping phenotypes from scientific literature or patient clinical records to the Human Phenotype Ontology. Recent advances in deep learning offer new opportunities for automation. We developed PhenoXtract, a novel phenotype extraction approach that combines Large Language Models and Knowledge Graph embedding. PhenoXtract is a multistep pipeline that takes clinical descriptions as input, extracts candidate phenotype entities using large language models, and maps them to terms from an enriched version of the Human Phenotype Ontology, processed as a knowledge graph. Results: Evaluation against expert-curated ground-truth datasets show a recall of 0.70 and precision of 0.85 for PhenoXtract, demonstrating concordance with manually extracted phenotypes, with a computation time of 10-20 seconds for each text analyzed. Moreover, PhenoXtract surpasses rule-based and deep learning-based state-of-the-art tools in two out of the three ground-truth datasets evaluated. These results suggest that hybrid approaches combining Large Language Models and Knowledge Graph embeddings represent a promising direction for automated clinical phenotyping at scale.

  • Methodological developments for omic data integration : applications to oncology and neurosciences

    by Galadriel Briere

    Développements méthodologiques pour l'intégration de données omiques : applications à l'oncologie et aux neurosciences: Les données dites "omiques", sont des données massives et hétérogènes, issues de la mesure de différents objets biologiques. Par exemple, la génomique s'intéresse à l'étude du génome (ADN), la transcriptomique à l'étude des transcrits (ARNs), la protéomique à l'étude des protéines, etc. L'interaction de l'ensemble de ces omiques entre elles ainsi qu'avec des facteurs environnementaux produit - à l’échelle d’une cellule, d’un tissu, ou d’un organisme - un ensemble de caractères observables appelé phénotype. Un des objectifs ultimes de la recherche en sciences de la vie est l’élucidation de la diversité du phénome (c’est-à-dire de l’ensemble des phénotypes observables) par l’identification des facteurs internes, environnementaux et de leurs interactions, associés à chaque phénotype.Ce manuscrit de thèse aborde la question de l’intégration de données - définie comme une solution permettant l'utilisation de multiples sources d'information (données) pour mieux comprendre un système, une situation, une association, etc. - et particulièrement de la question de l’intégration de données omiques, c’est-à-dire tout type d'intégration de sources de données provenant de différentes omiques, et/ou d'une même omique mesurée dans différents contextes expérimentaux et/ou de données omiques avec un type de données non-omique.Dans une première contribution, nous proposons une nouvelle stratégie pour le clustering consensus de données multi-omiques pour la détection de sous-types moléculaires de cancers. Cette stratégie permet, à partir de clusterings de cohortes de patients obtenus en considérant diverses données omiques et/ou différents algorithmes de clusterings existants, de produire un clustering consensus en réconciliant l’ensemble des prédictions contenues dans les clusterings soumis en entrée de l’algorithme. Deux scénarios d’intégration ont été testés : une intégration dite "multi-to-multi", produite par intégration de clusterings multi-omiques et une intégration dite "single-to-multi", produite par l’intégration de clusterings générés indépendamment pour différents omiques.Dans une seconde contribution, nous proposons une stratégie de détection de groupes de liens différentiellement co-exprimés identifiés par la comparaison de plusieurs jeux de données de type cas/contrôle. Elle repose sur la construction et l’analyse de réseaux multi-couches de co-expression différentielle, chaque couche représentant l'ensemble des dérégulations de la co-expression génique observée pour un contexte expérimental donné. La détection de groupes de liens de co-expression différentielle topologiquement similaires (c’est-à-dire impliquant un même ensemble de gènes) et observées dans les mêmes sous-ensembles de couches du réseau permet d’identifier des mécanismes associés à une maladie dans différents contextes expérimentaux (tissus, stade de développement, etc.), ou associés à différentes maladies. Nous avons appliqué la stratégie développée à la détection de motifs de co-expression différentielle dans l’hippocampe et le cortex de souris modèles de la maladie d’Alzheimer, ce qui nous a permis d’identifier des motifs clés de dérégulation de l’expression génique associés au phénotype pathologique. Certains de ces motifs ont été observés dans le cortex comme dans l’hippocampe, tandis que d’autres apparaissent spécifiques à l’une ou l’autre des deux structures cérébrales. Cette preuve de concept démontre la pertinence de la stratégie pour l’identification de perturbations de la co-régulation génique et la caractérisation transcriptomique de la diversité des phénotypes.

  • Open Access

    Lipidomics Identifies HFpEF Phenogroups and a High-Risk Metabolic Signature - The BElgian and CAnadian MEtabolomics in HFpEF (BECAME-HF) project

    by Julie Hussin, Nassiba Menghoum, Anik Forest, Pamela Mehanna, Olivier Tastet, Julie Thompson Legault, Isabelle Frayne Robillard, Sibille Lejeune, David Vancraeynest, Clotilde Roy, et al.

    Rationale: Heart failure with preserved ejection fraction (HFpEF) is a heterogeneous syndrome with substantial unmet diagnostic and therapeutic needs. Circulating lipid metabolism is increasingly implicated in HFpEF pathophysiology but has not been systematically leveraged for molecular stratification. Objective: To determine whether plasma lipidomics can identify molecular phenogroups of HFpEF associated with distinct clinical characteristics and outcomes. Methods and Results: Untargeted plasma lipidomics was performed in non-HF subjects and HFpEF patients from a primary Belgian cohort and an independent Canadian cohort (n=177 in each cohort). In the Belgian cohort, 235 unique lipids spanning 19 subclasses were annotated, including 96 significantly associated with HFpEF ( q <0.02). Unsupervised analyses revealed marked lipidomic heterogeneity, with a distinct HFpEF subgroup separable from non-HF subjects. Hierarchical clustering identified three phenogroups with divergent lipid profiles and clinical features. One phenogroup exhibited severe atrial dysfunction, congestion-related biomarkers, elevated indices of cardiac and liver fibrosis, and markedly reduced survival, a second was characterized by prominent metabolic syndrome features, and a third by preserved renal function. Cross-cohort comparison using a supervised classifier trained on 158 shared lipids confirmed analogous lower-risk phenogroups in the Canadian cohort, while the high-risk phenotype was underrepresented. A signature of 10 lipids across six subclasses, including long-chain acylcarnitines, ether phosphatidylcholines, and oxidized sphingomyelins, discriminated the high-risk group and correlated with markers of disease severity. Conclusion: Our findings demonstrate that HFpEF comprises metabolically distinct patient subgroups across cohorts, revealing specific lipidomic dysfunctions that deepen our understanding of HFpEF heterogeneity and underlying pathophysiology.

  • Open Access

    Using Networks and Prior Knowledge to Uncover novel Rare Disease Phenotypes

    by Galadriel Brière, Cécile Beust, Morgane Térézol, Anaïs Baudot

    Rare diseases are characterized by low prevalence and high phenotypic diversity. Accurately identifying phenotypes associated with rare diseases is crucial for facilitating their diagnosis and management. However, this task presents significant challenges: rare disease datasets are typically small, making statistical assessments difficult, and they often report phenotypes using various terminologies, hindering the identification of common phenotypes across datasets or the recognition of those already documented in literature and knowledge databases. The Xcelerate RARE 2023 challenge was established to address the identification of phenotypes associated with rare diseases. Our team, MAGNET, developed a network-based approach that integrates patient clinical data from the Xcelerate RARE 2023 challenge with existing knowledge from Orphanet and the Human Phenotype Ontology (HPO). Our approach first builds a patient-disease-phenotype network comprising two layers: the Xcelerate layer encoding disease-patient-symptom associations, and the Prior Knowledge layer incorporating relationships between Orphanet rare diseases and HPO phenotypes. Then, for each rare disease included in the Xcelerate dataset, a Random Walk with Restart (RWR) algorithm is applied to the multilayer network to prioritize phenotype nodes. This framework effectively prioritizes phenotypes associated with rare diseases while distinguishing novel phenotypes from those already documented in knowledge bases, hence offering new perspectives for improving the diagnosis and characterization of rare diseases. Our solution was awarded the prize for the most innovative approach in the Xcelerate RARE challenge.


  • Open Access

    Benchmarking the Impact of Data Leakage on the Performance of Knowledge Graph Embedding Models for Biomedical Link Prediction

    by Galadriel Brière, Thomas Stosskopf, Benjamin Loire, Anaïs Baudot

    In recent years, Knowledge Graphs (KGs) have received increasing attention for their ability to organize complex biomedical knowledge into structured representations of entities and relations. Knowledge Graph Embedding (KGE) models facilitate efficient exploration of KGs by learning compact data representations. These models are increasingly applied to biomedical KGs for link prediction, for instance to uncover new therapeutic uses for existing drugs. While numerous KGE models have been developed and benchmarked for link prediction, existing evaluations often over-look the critical issue of data leakage. Data leakage leads the model to learn patterns it would not encounter when deployed in real-world settings, artificially inflating performance metrics and compromising the overall validity of benchmark results. In machine learning, data leakage can arise when (1) there is redundancy between the training and test sets, (2) the model leverages illegitimate features, or (3) the test set does not accurately reflect real-world inference scenarios. In this study, we demonstrate the impact of train-test redundancies in KGE-based link prediction and implement a systematic control procedure to detect and remove those redundancies. In addition, through permutation experiments, we investigate whether node degree acts as an illegitimate predictive feature, and find no evidence that models rely on it. Finally, we evaluate how well common test set sampling strategies reflect the challenges of real-world inference in drug repurposing. We compare random and cold-start data splits with a real-world inference set derived from the Orphanet database. We find that performance drops significantly on the real-world inference set, indicating that current benchmarking practices may overestimate how well KGE models generalize to practical applications. Overall, our findings highlight the importance of rigorous benchmark design and careful evaluation of the generalization ability of KGE models for biomedical link prediction.


  • Open Access

    The Molecular Landscape of Premature Aging Diseases Defined by Multilayer Network Exploration

    by Cécile Beust, Alberto Valdeolivas, Anthony Baptista, Galadriel Brière, Nicolas Lévy, Ozan Ozisik, Anaïs Baudot

    Premature Aging (PA) diseases are rare genetic disorders that mimic some aspects of physiological aging at an early age. Various causative genes of PA diseases have been identified in recent years, providing insights into some dysfunctional cellular processes. However, the identification of PA genes also revealed significant genetic heterogeneity and highlighted the gaps in this understanding of PA‐associated molecular mechanisms. Furthermore, many patients remain undiagnosed. Overall, the current lack of knowledge about PA diseases hinders the development of effective diagnosis and therapies and poses significant challenges to improving patient care.Here, a network‐based approach to systematically unravel the cellular functions disrupted in PA diseases is presented. Leveraging a network community identification algorithm, it is delved into a vast multilayer network of biological interactions to extract the communities of 67 PA diseases from their 132 associated genes. It is found that these communities can be grouped into six distinct clusters, each reflecting specific cellular functions: DNA repair, cell cycle, transcription regulation, inflammation, cell communication, and vesicle‐mediated transport. That these clusters collectively represent the landscape of the molecular mechanisms that are perturbed in PA diseases, providing a framework for better understanding their pathogenesis is proposed. Intriguingly, most clusters also exhibited a significant enrichment in genes associated with physiological aging, suggesting a potential overlap between the molecular underpinnings of PA diseases and natural aging.


  • Open Access

    The molecular landscape of premature aging diseases defined by multilayer network exploration

    by Cécile Beust, Alberto Valdeolivas, Anthony Baptista, Galadriel Brière, Nicolas Lévy, Ozan Ozisik, Anaïs Baudot

    Premature Aging (PA) diseases are rare genetic disorders that mimic some aspects of physiological aging at an early age. Various causative genes of PA diseases have been identified in recent years, providing insights into some dysfunctional cellular processes. However, the identification of PA genes also revealed significant genetic heterogeneity and highlighted the gaps in our understanding of PA-associated molecular mechanisms. Furthermore, many patients remain undiagnosed. Overall, the current lack of knowledge about PA diseases hinders the development of effective diagnosis and therapies and poses significant challenges to improving patient care. Here, we present a network-based approach to systematically unravel the cellular functions disrupted in PA diseases. Leveraging a network community identification algorithm, we delved into a vast multilayer network of biological interactions to extract the communities of 67 PA diseases from their 132 associated genes. We found that these communities can be grouped into six distinct clusters, each reflecting specific cellular functions: DNA repair, cell cycle, transcription regulation, inflammation, cell communication, and vesicle-mediated transport. We propose that these clusters collectively represent the landscape of the molecular mechanisms that are perturbed in PA diseases, providing a framework for better understanding their pathogenesis. Intriguingly, most clusters also exhibited a significant enrichment in genes associated with physiological aging, suggesting a potential overlap between the molecular underpinnings of PA diseases and natural aging.

  • Open Access

    Random Walk with Restart on multilayer networks: from node prioritisation to supervised link prediction and beyond

    by Anthony Baptista, Galadriel Brière, Anaïs Baudot

    Background: Biological networks have proven invaluable ability for representing biological knowledge. Multilayer networks, which gather different types of nodes and edges in multiplex, heterogeneous and bipartite networks, provide a natural way to integrate diverse and multi-scale data sources into a common framework. Recently, we developed MultiXrank, a Random Walk with Restart algorithm able to explore such multilayer networks. MultiXrank outputs scores reflecting the proximity between an initial set of seed node(s) and all the other nodes in the multilayer network. We illustrate here the versatility of bioinformatics tasks that can be performed using MultiXrank. Results: We first show that MultiXrank can be used to prioritise genes and drugs of interest by exploring multilayer networks containing interactions between genes, drugs, and diseases. In a second study, we illustrate how MultiXrank scores can also be used in a supervised strategy to train a binary classifier to predict gene-disease associations. The classifier performance are validated using outdated and novel gene-disease association for training and evaluation, respectively. Finally, we show that MultiXrank scores can be used to compute diffusion profiles and use them as disease signatures. We computed the diffusion profiles of more than 100 immune diseases using a multilayer network that includes cell-type specific genomic information. The clustering of the immune disease diffusion profiles reveals shared shared phenotypic characteristics. Conclusion: Overall, we illustrate here diverse applications of MultiXrank to showcase its versatility. We expect that this can lead to further and broader bioinformatics applications.


  • Open Access

    Consensus clustering applied to multi-omics disease subtyping

    by Galadriel Brière, Élodie Darbo, Patricia Thébault, Raluca Uricaru

    Background: Facing the diversity of omics data and the difficulty of selecting one result over all those produced by several methods, consensus strategies have the potential to reconcile multiple inputs and to produce robust results. Results: Here, we introduce ClustOmics, a generic consensus clustering tool that we use in the context of cancer subtyping. ClustOmics relies on a non-relational graph database, which allows for the simultaneous integration of both multiple omics data and results from various clustering methods. This new tool conciliates input clusterings, regardless of their origin, their number, their size or their shape. ClustOmics implements an intuitive and flexible strategy, based upon the idea of evidence accumulation clustering. ClustOmics computes co-occurrences of pairs of samples in input clusters and uses this score as a similarity measure to reorganize data into consensus clusters. Conclusion: We applied ClustOmics to multi-omics disease subtyping on real TCGA cancer data from ten different cancer types. We showed that ClustOmics is robust to heterogeneous qualities of input partitions, smoothing and reconciling preliminary predictions into high-quality consensus clusters, both from a computational and a biological point of view. The comparison to a state-of-the-art consensus-based integration tool, COCA, further corroborated this statement. However, the main interest of ClustOmics is not to compete with other tools, but rather to make profit from their various predictions when no gold-standard metric is available to assess their significance. Availability: The ClustOmics source code, released under MIT license, and the results obtained on TCGA cancer data are available on GitHub: https://github.com/galadrielbriere/ClustOmics.