Claire Dandine-Roulland
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
Research Subjects
Claire Dandine-Roulland has a total of 97 co-authors. Here are the top co-authors by number of shared publications:
- Hervé Perdry
Université Paris-Sud
3 shared publications - Cathy Philippe
UNATI, Neurospin, Institut Joliot, CEA, Universite Paris-Saclay, Gif-sur-Yvette 91191, France
2 shared publications - Edith Le Floch
Institut François Jacob CEA Université Paris-Saclay Centre National de Recherche en Génomique Humaine
2 shared publications Jean-François Deleuze
Centre National de Recherche en Génomique Humaine, Institut de Biologie François Jacob, CEA, Université Paris-Saclay, Evry, France
2 shared publications- Jean-François Deleuze
Commissariat à l'Energie Atomique et aux Energies Alternatives
2 shared publications - Jean-François Mangin
Université Paris-Saclay
2 shared publications
Latest Publications
- 1 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Modélisation de la composante génétique des maladies humaines : Données familiales et Modèles Mixtes
Le modèle linéaire mixte a été formalisé il y a plus de 60 ans. Celui-ci permet d'estimer un modèle avec des effets fixes équivalents à ceux du modèle linéaire classique et des effets aléatoires. Ce type de modélisation, d'abord utilisé en génétique animale, est depuis quelques années largement utilisé en génétique humaine. Les utilisations de ce modèle sont nombreuses. En effet, il peut être utilisé en étude de liaison, d'association, pour l'estimation de l'héritabilité ou encore dans la recherche d'empreinte parentale et peut s'adapter à des données familiales ou en population.Le but de mon doctorat est d'exploiter différentes méthodes basées sur les modèles mixtes d'abord sur des données génétiques en population puis sur des données génétiques familiales.Dans un premier temps, nous explorons dans ce manuscrit la théorie des modèles linéaires mixtes et leur utilisation en génétique. Nous adaptons aussi certaines méthodes pour les appliquer à notre recherche. Ce travail a donné lieu au développement informatique d'un package R permettant d'utiliser ces modèles dans le cadre des études génétiques.Dans un deuxième temps, nous utilisons les modèles linéaires mixtes pour l'estimation de l'héritabilité dans une étude en population française, l'étude Trois-Cités. Nous disposons dans cette étude des génotypes des tag-SNPs habituellement utilisés dans les études d'association ainsi que des lieux de naissance et de plusieurs traits anthropométriques quantitatifs tels que la taille. L'objectif est alors d'étudier la présence et la prise en compte dans l'analyse de stratification de population dans cette étude. Dans ce manuscrit, nous analysons les coordonnées géographiques des lieux de naissance. Nos résultats mettent en évidence la difficulté pour corriger correctement la stratification de population avec les méthodes classiques dans certains cas. Nous analysons ensuite les traits anthropométriques en particulier la taille dont nous estimons l'héritabilité à 39% dans la population de l'étude Trois-Cités.Dans la dernière partie de ce manuscrit, nous nous concentrons sur les données familiales. Nous montrons le gain d’information que peut apporter ce type de données dans la recherche des variants causaux. Puis, nous explorons l'utilisation des modèles mixtes sur des données familiales en appliquant certaines des méthodes associées dans la recherche de signaux d'association pour la Sclérose en Plaques, une maladie auto-immune, en utilisant un échantillon d’une centaine de familles nucléaires avec au moins deux germains atteints. Nous avons alors mis en évidence l’inadéquation des méthodes classiques basées sur les modèles mixtes à ce type de données. Afin de mieux comprendre ce biais de sélection et de le corriger, plus d’investigations sont nécessaires.
- 1 April 2026
- Published by Agence Bibliographique de l'Enseignement Supérieur
Modélisation de la composante génétique des maladies humaines : Données familiales et Modèles Mixtes
Le modèle linéaire mixte a été formalisé il y a plus de 60 ans. Celui-ci permet d'estimer un modèle avec des effets fixes équivalents à ceux du modèle linéaire classique et des effets aléatoires. Ce type de modélisation, d'abord utilisé en génétique animale, est depuis quelques années largement utilisé en génétique humaine. Les utilisations de ce modèle sont nombreuses. En effet, il peut être utilisé en étude de liaison, d'association, pour l'estimation de l'héritabilité ou encore dans la recherche d'empreinte parentale et peut s'adapter à des données familiales ou en population.Le but de mon doctorat est d'exploiter différentes méthodes basées sur les modèles mixtes d'abord sur des données génétiques en population puis sur des données génétiques familiales.Dans un premier temps, nous explorons dans ce manuscrit la théorie des modèles linéaires mixtes et leur utilisation en génétique. Nous adaptons aussi certaines méthodes pour les appliquer à notre recherche. Ce travail a donné lieu au développement informatique d'un package R permettant d'utiliser ces modèles dans le cadre des études génétiques.Dans un deuxième temps, nous utilisons les modèles linéaires mixtes pour l'estimation de l'héritabilité dans une étude en population française, l'étude Trois-Cités. Nous disposons dans cette étude des génotypes des tag-SNPs habituellement utilisés dans les études d'association ainsi que des lieux de naissance et de plusieurs traits anthropométriques quantitatifs tels que la taille. L'objectif est alors d'étudier la présence et la prise en compte dans l'analyse de stratification de population dans cette étude. Dans ce manuscrit, nous analysons les coordonnées géographiques des lieux de naissance. Nos résultats mettent en évidence la difficulté pour corriger correctement la stratification de population avec les méthodes classiques dans certains cas. Nous analysons ensuite les traits anthropométriques en particulier la taille dont nous estimons l'héritabilité à 39% dans la population de l'étude Trois-Cités.Dans la dernière partie de ce manuscrit, nous nous concentrons sur les données familiales. Nous montrons le gain d’information que peut apporter ce type de données dans la recherche des variants causaux. Puis, nous explorons l'utilisation des modèles mixtes sur des données familiales en appliquant certaines des méthodes associées dans la recherche de signaux d'association pour la Sclérose en Plaques, une maladie auto-immune, en utilisant un échantillon d’une centaine de familles nucléaires avec au moins deux germains atteints. Nous avons alors mis en évidence l’inadéquation des méthodes classiques basées sur les modèles mixtes à ce type de données. Afin de mieux comprendre ce biais de sélection et de le corriger, plus d’investigations sont nécessaires.
- Open Access
- 8 September 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (9)
GNNenrich: a novel method for pathway enrichment analysis based on Graph Neural Network
Motivation: Graph Neural Network (GNN) models have emerged in many fields and notably for biological networks constituted by genes or proteins and their interactions. The majority of enrichment study methods apply over-representation analysis and gene/protein set scores according to the existing overlap between pathways. Such methods neglect knowledges coming from the interactions between the gene/protein sets. Here, we introduce a novel GNN-based enrichment analysis method called GNNenrich. GNNenrich, through multiple levels of embedding that integrate protein sequence properties and interactions network, establishes functional relationship to support biological interpretation. Results: GNNenrich have been tested and compared to Over-Representation Analysis technique (g: Profiler) and graph-based method (EnrichNet). It demonstrates the capacity to reproduce results provided by others approaches and offers new perspectives for interpretation, returning relevant results supported by Protein–Protein Interactions (PPIs). Availability and implementation: Source code is available at https://gitlab.com/cnrgh/gnn-enrich/gnn-enrich-article-demo.
- 8 September 2025
- Published by Oxford University Press (OUP) in Bioinformatics
- Vol. 41 (9)
Open AccessGNNenrich: a novel method for pathway enrichment analysis based on Graph Neural Network
Motivation: Graph Neural Network (GNN) models have emerged in many fields and notably for biological networks constituted by genes or proteins and their interactions. The majority of enrichment study methods apply over-representation analysis and gene/protein set scores according to the existing overlap between pathways. Such methods neglect knowledges coming from the interactions between the gene/protein sets. Here, we introduce a novel GNN-based enrichment analysis method called GNNenrich. GNNenrich, through multiple levels of embedding that integrate protein sequence properties and interactions network, establishes functional relationship to support biological interpretation. Results: GNNenrich have been tested and compared to Over-Representation Analysis technique (g: Profiler) and graph-based method (EnrichNet). It demonstrates the capacity to reproduce results provided by others approaches and offers new perspectives for interpretation, returning relevant results supported by Protein–Protein Interactions (PPIs). Availability and implementation: Source code is available at https://gitlab.com/cnrgh/gnn-enrich/gnn-enrich-article-demo.
- Open Access
- 13 November 2022
- Published by Elsevier in European Journal of Cancer
- Vol. 179, 76-86
Association and performance of polygenic risk scores for breast cancer among French women presenting or not a familial predisposition to the disease
- 13 November 2022
- Published by Elsevier in European Journal of Cancer
- Vol. 179, 76-86
Open AccessAssociation and performance of polygenic risk scores for breast cancer among French women presenting or not a familial predisposition to the disease
- Open Access
- 4 March 2021
- Published by Springer Nature in European Journal of Human Genetics
- Vol. 29 (9) , 1424-1437
Genome-wide haplotype association study in imaging genetics using whole-brain sulcal openings of 16,304 UK Biobank subjects
Neuroimaging-genetics cohorts gather two types of data: brain imaging and genetic data. They allow the discovery of associations between genetic variants and brain imaging features. They are invaluable resources to study the influence of genetics and environment in the brain features variance observed in normal and pathological populations. This study presents a genome-wide haplotype analysis for 123 brain sulcus opening value (a measure of sulcal width) across the whole brain that include 16,304 subjects from UK Biobank. Using genetic maps, we defined 119,548 blocks of low recombination rate distributed along the 22 autosomal chromosomes and analyzed 1,051,316 haplotypes. To test associations between haplotypes and complex traits, we designed three statistical approaches. Two of them use a model that includes all the haplotypes for a single block, while the last approach considers each haplotype independently. All the statistics produced were assessed as rigorously as possible. Thanks to the rich imaging dataset at hand, we used resampling techniques to assess False Positive Rate for each statistical approach in a genome-wide and brain-wide context. The results on real data show that genome-wide haplotype analyses are more sensitive than single-SNP approach and account for local complex Linkage Disequilibrium (LD) structure, which makes genome-wide haplotype analysis an interesting and statistically sound alternative to the single-SNP counterpart.
- 4 March 2021
- Published by Springer Nature in European Journal of Human Genetics
- Vol. 29 (9) , 1424-1437
Open AccessGenome-wide haplotype association study in imaging genetics using whole-brain sulcal openings of 16,304 UK Biobank subjects
Neuroimaging-genetics cohorts gather two types of data: brain imaging and genetic data. They allow the discovery of associations between genetic variants and brain imaging features. They are invaluable resources to study the influence of genetics and environment in the brain features variance observed in normal and pathological populations. This study presents a genome-wide haplotype analysis for 123 brain sulcus opening value (a measure of sulcal width) across the whole brain that include 16,304 subjects from UK Biobank. Using genetic maps, we defined 119,548 blocks of low recombination rate distributed along the 22 autosomal chromosomes and analyzed 1,051,316 haplotypes. To test associations between haplotypes and complex traits, we designed three statistical approaches. Two of them use a model that includes all the haplotypes for a single block, while the last approach considers each haplotype independently. All the statistics produced were assessed as rigorously as possible. Thanks to the rich imaging dataset at hand, we used resampling techniques to assess False Positive Rate for each statistical approach in a genome-wide and brain-wide context. The results on real data show that genome-wide haplotype analyses are more sensitive than single-SNP approach and account for local complex Linkage Disequilibrium (LD) structure, which makes genome-wide haplotype analysis an interesting and statistically sound alternative to the single-SNP counterpart.
- Open Access
Genome-wide haplotype association study in imaging genetics using whole-brain sulcal openings of 16,304 UK Biobank subjects
Neuroimaging-genetics cohorts gather two types of data: brain imaging and genetic data. They allow the discovery of associations between genetic variants and brain imaging features. They are invaluable resources to study the influence of genetics and environment in the brain features variance observed in normal and pathological populations. This study presents a genome wide haplotype analysis for 123 brain sulcus opening value (a measure of sulcal width) across the whole brain that include 16,304 subjects from UK Biobank. Using genetic maps, we defined 119,548 blocks of low recombination rate distributed along the 22 autosomal chromosomes, and analyzed 1,051,316 haplotypes. To test associations between haplotypes and complex traits, we designed three statistical approaches. Two of them use a model that includes all the haplotypes for a sin gle block, while the last approach considers one model by haplotype. All the statistics produced were assessed as rigorously as possible. Thanks to the rich imaging dataset at hand, we used resampling techniques to assess False Positive Rate for each statistical approach in a genome-wide and brain-wide context. The results on real data show that genome-wide haplotype analyses are more sensitive than single-SNP approach and account for local complex Linkage Disequilibrium (LD) structure, which makes genome-wide haplotype analysis an interesting and statistically sound alternative to the single-SNP counterpart.
Open AccessGenome-wide haplotype association study in imaging genetics using whole-brain sulcal openings of 16,304 UK Biobank subjects
Neuroimaging-genetics cohorts gather two types of data: brain imaging and genetic data. They allow the discovery of associations between genetic variants and brain imaging features. They are invaluable resources to study the influence of genetics and environment in the brain features variance observed in normal and pathological populations. This study presents a genome wide haplotype analysis for 123 brain sulcus opening value (a measure of sulcal width) across the whole brain that include 16,304 subjects from UK Biobank. Using genetic maps, we defined 119,548 blocks of low recombination rate distributed along the 22 autosomal chromosomes, and analyzed 1,051,316 haplotypes. To test associations between haplotypes and complex traits, we designed three statistical approaches. Two of them use a model that includes all the haplotypes for a sin gle block, while the last approach considers one model by haplotype. All the statistics produced were assessed as rigorously as possible. Thanks to the rich imaging dataset at hand, we used resampling techniques to assess False Positive Rate for each statistical approach in a genome-wide and brain-wide context. The results on real data show that genome-wide haplotype analyses are more sensitive than single-SNP approach and account for local complex Linkage Disequilibrium (LD) structure, which makes genome-wide haplotype analysis an interesting and statistically sound alternative to the single-SNP counterpart.
- Open Access
- 20 March 2019
- Published by The Royal Society in Proceedings Of The Royal Society B-Biological Sciences
- Vol. 286 (1899) , 20182664
Genomic evidence for MHC disassortative mating in humans
Although pervasive in many animal species, the evidence for major histocompatibility complex (MHC) disassortative mating in humans remains inconsistent across studies. Here, to revisit this issue, we analyse dense genotype data for 883 European and Middle Eastern couples. To distinguish MHC-specific effects from socio-cultural confounders, the pattern of relatedness between spouses in the MHC region is compared to the rest of the genome. Couples from Israel exhibit no significant pattern of relatedness across the MHC region, whereas across the genome, they are more similar than random pairs of individuals, which may reflect social homogamy and/or cousin marriages. On the other hand, couples from The Netherlands and more generally from Northern Europe are significantly more MHC-dissimilar than random pairs of individuals, and this pattern of dissimilarity is extreme when compared with the rest of the genome. Our findings support the hypothesis that the MHC influences mate choice in humans in a context-dependent way: MHC-driven preferences may exist in all populations but, in some populations, social constraints over mate choice may reduce the ability of individuals to rely on such biological cues when choosing their mates.
- 20 March 2019
- Published by The Royal Society in Proceedings Of The Royal Society B-Biological Sciences
- Vol. 286 (1899) , 20182664
Open AccessGenomic evidence for MHC disassortative mating in humans
Although pervasive in many animal species, the evidence for major histocompatibility complex (MHC) disassortative mating in humans remains inconsistent across studies. Here, to revisit this issue, we analyse dense genotype data for 883 European and Middle Eastern couples. To distinguish MHC-specific effects from socio-cultural confounders, the pattern of relatedness between spouses in the MHC region is compared to the rest of the genome. Couples from Israel exhibit no significant pattern of relatedness across the MHC region, whereas across the genome, they are more similar than random pairs of individuals, which may reflect social homogamy and/or cousin marriages. On the other hand, couples from The Netherlands and more generally from Northern Europe are significantly more MHC-dissimilar than random pairs of individuals, and this pattern of dissimilarity is extreme when compared with the rest of the genome. Our findings support the hypothesis that the MHC influences mate choice in humans in a context-dependent way: MHC-driven preferences may exist in all populations but, in some populations, social constraints over mate choice may reduce the ability of individuals to rely on such biological cues when choosing their mates.
- Open Access
- 25 May 2016
- Published by Springer Nature in Scientific Reports
- Vol. 6 (1) , 26471
Accuracy of heritability estimations in presence of hidden population stratification
The heritability of a trait is the proportion of its variance explained by genetic factors; it has historically been estimated using familial data. However, new methods have appeared for estimating heritabilities using genomewide data from unrelated individuals. A drawback of this strategy is that population stratification can bias the estimates. Indeed, an environmental factor associated with the phenotype may differ among population subgroups. This factor being associated both with the phenotype and the genetic variation in the population would be a confounder. A common solution consists in adjusting on the first Principal Components (PCs) of the genomic data. We study this procedure on simulated data and on 6000 individuals from the Three-City Study. We analyse the geographical coordinates of the birth cities, which are not genetically determined, but the heritability of which should be overestimated due to population stratification. We also analyse various anthropometric traits. The procedure fails to correct the bias in geographical coordinates heritability estimates. The heritability estimates of the anthropometric traits are affected by the inclusion of the first PC, but not by the following PCs, contrarily to geographical coordinates. We recommend to be cautious with heritability estimates obtained from a large population.
- 25 May 2016
- Published by Springer Nature in Scientific Reports
- Vol. 6 (1) , 26471
Open AccessAccuracy of heritability estimations in presence of hidden population stratification
The heritability of a trait is the proportion of its variance explained by genetic factors; it has historically been estimated using familial data. However, new methods have appeared for estimating heritabilities using genomewide data from unrelated individuals. A drawback of this strategy is that population stratification can bias the estimates. Indeed, an environmental factor associated with the phenotype may differ among population subgroups. This factor being associated both with the phenotype and the genetic variation in the population would be a confounder. A common solution consists in adjusting on the first Principal Components (PCs) of the genomic data. We study this procedure on simulated data and on 6000 individuals from the Three-City Study. We analyse the geographical coordinates of the birth cities, which are not genetically determined, but the heritability of which should be overestimated due to population stratification. We also analyse various anthropometric traits. The procedure fails to correct the bias in geographical coordinates heritability estimates. The heritability estimates of the anthropometric traits are affected by the inclusion of the first PC, but not by the following PCs, contrarily to geographical coordinates. We recommend to be cautious with heritability estimates obtained from a large population.
- Open Access
- 14 January 2015
- Published by Springer Nature in European Journal of Human Genetics
- Vol. 23 (10) , 1357-1363
Where is the causal variant? On the advantage of the family design over the case–control design in genetic association studies
Many associated single-nucleotide polymorphisms (SNPs) have been identified by association studies for numerous diseases. However, the association between a SNP and a disease can result from a causal variant in linkage disequilibrium (LD) with the considered SNP. Assuming that the true causal variant is among the genotyped SNPs, other authors demonstrated that the power to discriminate between it and other SNPs in LD is low. Here, we propose to take advantage of the information provided by family data to improve the inference on the causal variant: we exploit the linkage information provided by affected sib pairs to discriminate the causal variant from the associated SNPs. The family-based approach improves discrimination power requiring up to five times less individuals than its case–control equivalent. However, the main advantage of family design is the possibility to carry out the procedure one step further: the linkage information allows inference on causal variants, which are not genotyped but in LD with tag-SNPs displaying association, which is impossible with case–control design. By means of Bayesian methods, we estimate the LD between the observed SNPs and an unobserved causal variant, as well as the allelic odds ratio at the unobserved causal variant. The proposed procedure is illustrated on a multiple sclerosis (MS) family data set including genotypes of SNPs in IL2RA, confirming the advantage of using a family design to identify causal variants. The results of our method on this data suggest the existence of two distinct causal variants in this gene for the MS.
- 14 January 2015
- Published by Springer Nature in European Journal of Human Genetics
- Vol. 23 (10) , 1357-1363
Open AccessWhere is the causal variant? On the advantage of the family design over the case–control design in genetic association studies
Many associated single-nucleotide polymorphisms (SNPs) have been identified by association studies for numerous diseases. However, the association between a SNP and a disease can result from a causal variant in linkage disequilibrium (LD) with the considered SNP. Assuming that the true causal variant is among the genotyped SNPs, other authors demonstrated that the power to discriminate between it and other SNPs in LD is low. Here, we propose to take advantage of the information provided by family data to improve the inference on the causal variant: we exploit the linkage information provided by affected sib pairs to discriminate the causal variant from the associated SNPs. The family-based approach improves discrimination power requiring up to five times less individuals than its case–control equivalent. However, the main advantage of family design is the possibility to carry out the procedure one step further: the linkage information allows inference on causal variants, which are not genotyped but in LD with tag-SNPs displaying association, which is impossible with case–control design. By means of Bayesian methods, we estimate the LD between the observed SNPs and an unobserved causal variant, as well as the allelic odds ratio at the unobserved causal variant. The proposed procedure is illustrated on a multiple sclerosis (MS) family data set including genotypes of SNPs in IL2RA, confirming the advantage of using a family design to identify causal variants. The results of our method on this data suggest the existence of two distinct causal variants in this gene for the MS.
- 1 January 2015
- Published by S. Karger AG in Human Heredity
- Vol. 80 (4) , 196-206
The Use of the Linear Mixed Model in Human Genetics
We give a short but detailed review of the methods used to deal with linear mixed models (restricted likelihood, AIREML algorithm, best linear unbiased predictors, etc.), with a few original points. Then we describe three common applications of the linear mixed model in contemporary human genetics: association testing (pathways analysis or rare variants association tests), genomic heritability estimates, and correction for population stratification in genome-wide association studies. We also consider the performance of best linear unbiased predictors for prediction in this context, through a simulation study for rare variants in a short genomic region, and through a short theoretical development for genome-wide data. For each of these applications, we discuss the relevance and the impact of modeling genetic effects as random effects.
- 1 January 2015
- Published by S. Karger AG in Human Heredity
- Vol. 80 (4) , 196-206
The Use of the Linear Mixed Model in Human Genetics
We give a short but detailed review of the methods used to deal with linear mixed models (restricted likelihood, AIREML algorithm, best linear unbiased predictors, etc.), with a few original points. Then we describe three common applications of the linear mixed model in contemporary human genetics: association testing (pathways analysis or rare variants association tests), genomic heritability estimates, and correction for population stratification in genome-wide association studies. We also consider the performance of best linear unbiased predictors for prediction in this context, through a simulation study for rare variants in a short genomic region, and through a short theoretical development for genome-wide data. For each of these applications, we discuss the relevance and the impact of modeling genetic effects as random effects.
- Open Access
- 1 December 2014
- Published by Wolters Kluwer Health in Medicine
- Vol. 93 (29) , e280
Environmental Influences on Daily Emergency Admissions in Sickle-Cell Disease Patients
Previous reports have suggested a role for weather conditions and air pollution on the variability of sickle cell disease (SCD) severity, but large-scale comprehensive epidemiological studies are lacking. In order to evaluate the influence of air pollution and climatic factors on emergency hospital admissions (EHA) in SCD patients, we conducted an 8-year observational retrospective study in 22 French university hospitals in Paris conurbation, using distributed lag non-linear models, a methodology able to flexibly describe simultaneously non-linear and delayed associations, with a multivariable approach. During the 2922 days of the study, there were 17,710 EHA, with a mean daily number of 6.1 ± 2.8. Most environmental factors were significantly correlated to each other. The risk of EHA was significantly associated with higher values of nitrogen dioxide, atmospheric particulate matters, and daily mean wind speed; and with lower values of carbon monoxide, ozone, sulfur dioxide, daily temperature (minimal, maximal, mean, and range), day-to-day mean temperature change, daily bright sunshine, and occurrence of storm. There was a lag effect for 12 of 15 environmental factors influencing hospitalization rate. Multivariate analysis identified carbon monoxide, day-to-day temperature change, and mean wind speed, along with calendar factors (weekend, summer season, and year) as independent factors associated with EHA. In conclusion, most weather conditions and air pollutants assessed were correlated to each other and influenced the rate of EHA in SCD patients. In multivariate analysis, lower carbon monoxide concentrations, day-to-day mean temperature drop and higher wind speed were associated with increased risk of EHA.
- 1 December 2014
- Published by Wolters Kluwer Health in Medicine
- Vol. 93 (29) , e280
Open AccessEnvironmental Influences on Daily Emergency Admissions in Sickle-Cell Disease Patients
Previous reports have suggested a role for weather conditions and air pollution on the variability of sickle cell disease (SCD) severity, but large-scale comprehensive epidemiological studies are lacking. In order to evaluate the influence of air pollution and climatic factors on emergency hospital admissions (EHA) in SCD patients, we conducted an 8-year observational retrospective study in 22 French university hospitals in Paris conurbation, using distributed lag non-linear models, a methodology able to flexibly describe simultaneously non-linear and delayed associations, with a multivariable approach. During the 2922 days of the study, there were 17,710 EHA, with a mean daily number of 6.1 ± 2.8. Most environmental factors were significantly correlated to each other. The risk of EHA was significantly associated with higher values of nitrogen dioxide, atmospheric particulate matters, and daily mean wind speed; and with lower values of carbon monoxide, ozone, sulfur dioxide, daily temperature (minimal, maximal, mean, and range), day-to-day mean temperature change, daily bright sunshine, and occurrence of storm. There was a lag effect for 12 of 15 environmental factors influencing hospitalization rate. Multivariate analysis identified carbon monoxide, day-to-day temperature change, and mean wind speed, along with calendar factors (weekend, summer season, and year) as independent factors associated with EHA. In conclusion, most weather conditions and air pollutants assessed were correlated to each other and influenced the rate of EHA in SCD patients. In multivariate analysis, lower carbon monoxide concentrations, day-to-day mean temperature drop and higher wind speed were associated with increased risk of EHA.