Skip to main content

ENFOQUE MACHINE LEARNING Y BIG DATA MULTI-OMICO: MEDICINA PERSONALIZADA

PI20/00711

•
Funding agency name Instituto de Salud Carlos III
Funding agency acronym ISCIII
Program Programa Estatal de Generación de Conocimiento y Fortalecimiento del Sistema Español de I+D+I
Subprogram Subprograma Estatal de Generación de Conocimiento
Call Proyectos de Investigación en Salud
Call year 2020
State Plan 2017-2020 Plan Estatal de Investigación Científica y Técnica y de Innovación
Beneficiary centre UNIVERSIDAD DE GRANADA
Execution centre UNIVERSIDAD DE GRANADA

Resultados relacionados

Resultados totales (Incluyendo duplicados): 6
Encontrada(s) 1 página(s)

Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Contribución de congreso (ConferenceOutput)

Leveraging Machine Learning and Genetic Risk Scores for the Prediction of Metabolic Syndrome in Children with Obesity

  • Aguilera, Concepción M.
  • Bustos-Aibar, Mireia
  • Anguita-Ruiz, Augusto
  • Torres-Martos, Álvaro
  • Bueno, Gloria
  • Leis, Rosaura
  • Alcalá-Fernández, Jesús
Background and objectives: Obesity is a growing global epidemic, associated with increased cardiometabolic disorders. Metabolic syndrome (MS) is defined by altered insulin, blood pressure, glucose, and lipid levels. Pubertal children with obesity are highly susceptible to developing MS, necessitating its early identification. This study aims to compute phenotype-specific genetic risk scores for MS-related biochemical markers and evaluate their clinical utility using machine learning- based models. Methods: Longitudinal data from the PUBMEP Spanish cohort were analyzed, including 138 children (71 girls and 67 boys) at two time points, spanning from prepuberty to puberty. Clinical, endogenous, environmental, and omics variables were measured. Genetic risk scores were generated using GWAS data and PRSice-2 software. These scores, alongside non-genetic prepubertal data (e.g., biochemical, anthropometric, and physical activity data), were integrated into predictive models using machine learning techniques to forecast the MS status during puberty. Linear models explored interactions between environmental factors, genetic risk scores, and disease risk. Results: Strong associations were observed between each genetic risk score and its corresponding phenotypic biomarker. Notably, certain scores related to obesity and high-density lipoprotein levels exhibited significant interactions with environmental factors, such as sedentary lifestyle, modulating disease effects. The predictive machine learning models incorporating prepubertal genetics, high- density lipoprotein, and sedentary lifestyle achieved reasonable performance in predicting pubertal obesity (AUC, accuracy, and sensitivity of 0.89). These models strike a favorable balance between risk scores derived from genetic factors and clinical variables. However, when individual risk scores were considered in isolation, limited predictive results were observed for MS and associated altered components. Discussion: This study demonstrates the importance of the early identification of at-risk children for MS. The integration of genetic risk scores, clinical variables, and machine learning techniques offers promising avenues for predicting pubertal MS. While individual risk scores have limitations in isolation, polygenic risk scores serve as valuable tools for investigating gene– environment interactions. Following our results, polygenic risk scores lacked sufficient predictive ability in most clinical traits, limiting their clinical application. Nevertheless, they remain valuable analytical tools for exploring the association with the environment, by consolidating the effects of multiple single nucleotide polymorphisms into a single variable.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Multiomics and eXplainable artificial intelligence for decision support in insulin resistance early diagnosis: a pediatric population-based longitudinal study

  • Torres-Martos, Álvaro
  • Anguita-Ruiz, Augusto
  • Bustos-Aibar, Mireia
  • Ramírez-Mena, Alberto
  • Arteaga, María
  • Bueno, Gloria
  • Leis, Rosaura
  • Aguilera, Concepción M.
  • Alcalá, Rafael
  • Alcalá-Fdez, Jesús
Pediatric obesity can drastically heighten the risk of cardiometabolic alterations later in life, with insulin resistance standing as the cornerstone linking adiposity to the increased cardiovascular risk. Puberty has been pointed out as a critical stage after which obesity-associated insulin resistance is more difficult to revert. Timely prediction of insulin resistance in pediatric obesity is therefore vital for mitigating the risk of its associated comorbidities. The construction of effective and robust predictive systems for a complex health outcome like insulin resistance during the early stages of life demands the adoption of longitudinal designs for more causal inferences, and the integration of factors of varying nature involved in its onset. In this work, we propose an eXplainable Artificial Intelligence-based decision support pipeline for early diagnosis of insulin resistance in a longitudinal cohort of 90 children. For that, we leverage multi-omics (genomics and epigenomics) and clinical data from the pre-pubertal stage. Different data layers combinations, pre-processing techniques (missing values, feature selection, class imbalance, etc.), algorithms, training procedures were considered following good practices for Machine Learning. SHapley Additive exPlanations were provided for specialists to understand both the decision-making mechanisms of the system and the impact of the features on each automatic decision, an essential issue in high-risk areas such as this one where system decisions may affect people’s lives. The system showed a relevant predictive ability (AUC and G-mean of 0.92). A deep exploration, both at the global and the local level, revealed promising biomarkers of insulin resistance in our population, highlighting classical markers, such as Body Mass Index z-score or leptin/adiponectin ratio, and novel ones such as methylation patterns of relevant genes, such as HDAC4, PTPRN2, MATN2, RASGRF1 and EBF1. Our findings highlight the importance of integrating multi-omics data and following eXplainable Artificial Intelligence trends when building decision support systems.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

ObMetrics: A Shiny app to assist in metabolic syndrome assessment in paediatric obesity

  • Torres-Martos, Álvaro
  • Requena, Francisco
  • López-Rodríguez, Guadalupe
  • Hernández-Cabrera, Jhazmin
  • Galván, Marcos
  • Solís-Pérez, Elizabeth
  • Romo-Tello, Susana
  • Jasso-Medrano, José Luis
  • Vilchis-Gil, Jenny
  • Klünder-Klünder, Miguel
  • Martínez-Andrade, Gloria
  • Enríquez, María Elena Acosta
  • Aristizabal, Juan Carlos
  • Ramírez-Mena, Alberto
  • Stratakis, Nikos
  • Bustos-Aibar, Mireia
  • Gil, Ángel
  • Gil-Campos, Mercedes
  • Bueno, Gloria
  • Leis, Rosaura
  • Alcalá-Fdez, Jesús
  • Aguilera, Concepción María
  • Anguita-Ruiz, Augusto
SummaryObjectiveTo introduce ObMetrics, a free and user‐friendly Shiny app that simplifies the calculation, data analysis, and interpretation of Metabolic Syndrome (MetS) outcomes according to multiple definitions in epidemiological studies of paediatric populations. We illustrate its usefulness using ethnically different populations in a comparative study of prevalence across cohorts and definitions.MethodsWe conducted a case study using data from two ethnically diverse paediatric populations: a Hispanic‐American cohort (N = 1759) and a Hispanic‐European cohort (N = 2411). Using ObMetrics, we computed MetS classifications (Cook, Zimmet, Ahrens) and component‐specific z‐scores for each participant to compare prevalences.ResultsThe analysis revealed significant heterogeneity in MetS prevalence across different definitions and cohorts. According to Cook, Zimmet, and Ahrens's definitions, MetS prevalence in children with obesity was 25%, 12%, and 48%, respectively, in the Hispanic‐European cohort, and 38%, 27%, and 66% in the Hispanic‐American cohort. Calculating component‐specific z‐scores in each cohort also highlighted ethnic‐specific differences in lipid metabolism and blood pressure. By automating these complex calculations, ObMetrics considerably reduced analysis time and minimised the potential for errors.ConclusionObMetrics proved to be a powerful tool for paediatric research, generating detailed reports on the prevalence of MetS and its components based on various definitions and reference standards. Our case study further provides valuable insights into the challenges of characterising metabolic health in paediatric populations. Future efforts should focus on developing unified consensus guidelines for paediatric MetS. Meanwhile, ObMetrics enables earlier identification and targeted intervention for high‐risk children and adolescents.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Sleep duration moderates association between screen time and emotional and behavioural problems in young children

  • Iglesias-Vázquez, Lucía
  • Garcidueñas-Fimbres, Tany E.
  • Gómez-Martínez, Carlos
  • Castro-Collado, Cristina
  • Leis, Rosaura
  • Fernández de la Puente, María
  • Moreno, Luis A.
  • Navas-Carretero, Santiago
  • Corella, Dolores
  • Moreira Echeverria, Ana
  • Jurado-Castro, José M.
  • Picáns-Leis, Rosaura
  • Ni, Jiaqi
  • Miguel-Berges, Maria L.
  • Martínez, J. Alfredo
  • Benedicto-Toboso, María I.
  • Llorente-Cantarero, Francisco
  • Vázquez-Cobela, Rocío
  • Feliu, Albert
  • Masip, Guiomar
  • Pastor-Villaescusa, Belén
  • Gil-Campos, Mercedes
  • Escribano, Joaquín
  • Babio, Nancy
Background Preschoolers and young children are vulnerable to psychosocial and behavioral disorders linked to lifestyle factors such as screen time and sleep disturbances. Our study examines the relationship between screen time and adherence to recommendations with children’s behavioral and emotional difficulties, with a focus on the role of sleep duration. Methods Cross-sectional analyses were conducted within the multicenter prospective Childhood Obesity Risk Assessment Longitudinal Study (CORALS), which included 1420 children aged 3–6 years. Screen time (hours/day) and adherence to recommendations (≤ 2 hours/day) were assessed. Behavioral and emotional difficulties were measured via the strengths and difficulties questionnaire. Multivariable linear and logistic regression models were used to estimate associations between screen time (continuous and dichotomous) and strengths and difficulties questionnaire scores, adjusting for potential confounders. We also tested the moderating effect of sleep and conducted isotemporal substitution analyses replacing screen time with sleep duration. Results Higher screen time was associated with higher total strengths and difficulties questionnaire scores [β 95% confidence interval (CI), 0.35 (0.10, 0.61)], emotional symptoms [0.10 (0.01, 0.19)], conduct problems [0.10 (0.01, 0.18)], and greater odds of exceeding the 16-point strengths and difficulties questionnaire cutoff for behavioral and emotional difficulties [odds ratio (OR) (95% CI), 1.21 (1.04, 1.41)]. Children who adhered to screen time recommendations had lower strengths and difficulties questionnaire total scores [β (95% CI), − 0.64 (− 1.19, − 0.10)] and odds of experiencing behavioral and emotional difficulties [OR (95% CI), 0.67 (0.47, 0.95)]. Sleep duration moderated the screen time–strengths and difficulties questionnaire association (P = 0.020). The isotemporal substitution of screen time for sleep duration was associated with lower strengths and difficulties questionnaire scores across all subscales, except for prosocial behavior. Conclusions Higher screen time was associated with greater emotional and behavioral difficulties, whereas adherence to screen time recommendations and adequate sleep duration were inversely associated. Managing screen time and promoting sleep are crucial for children’s well-being.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Novel epigenetic marks of insulin resistance trajectories in a longitudinal study of childhood obesity

  • Anguita-Ruiz, Augusto
  • Torres-Martos, Álvaro
  • Bustos-Aibar, Mireia
  • Setó-Llorens, Adrià
  • Ruiz-Ojeda, Francisco Javier
  • Moreno, Luis A.
  • Gil, Ángel
  • Gil-Campos, Mercedes
  • Bueno, Gloria
  • Leis, Rosaura
  • Alcalá-Fdez, Jesús
  • Aguilera, Concepción M.
Background Childhood obesity is a major global public-health challenge. Insulin resistance (IR) is a critical driver of later cardiometabolic alterations. A comprehensive understanding of the molecular mechanisms underlying the initial development of childhood IR is essential for timely prevention and intervention. In this study, we aimed to assess the association between IR and blood DNA methylation in a longitudinal study from childhood into adolescence. Methods The PUBMEP study included a longitudinal core of 90 children with paired blood samples collected at both pre-pubertal and pubertal stages. For cross-sectional analyses, this sample was expanded to 99 pre-pubertal and 129 pubertal participants. IR status was defined according to clinically relevant sex- and pubertal stage specific HOMA-IR cut-offs, as recommended by pediatric expert clinicians. Genotype data was obtained with the Infinium Global Screening Array, and blood DNA methylation sites with the Infinium MethylationEPIC BeadChip. Epigenome-wide associations with IR status and trajectories were tested using linear models in the longitudinal and cross-sectional sets. FDR-adjusted significant CpG sites were assessed with sex- and age-standardised cardiometabolic z-scores (adiposity, lipids, blood pressure, glycaemia and IR) at each stage. mQTL analyses were performed to identify genetic variants that drive IR-associated methylation signals. Results We identified 120 CpG sites related to obesity-associated IR in the context of pubertal transition that remained significant after global FDR correction (FDR < 0.05). These CpG sites showed distinct methylation profiles that tracked IR trajectories from prepuberty to puberty, with consistent differences across children whose IR improved, worsened or remained stable, with several of them also related to cardiometabolic traits at pubertal stage, including adiposity measures, blood pressure and glycaemic indices. Among the FDR-significant CpG sites with biological relevance for IR, methylation at CpG sites annotated to SLC2A9, PEPD, TSC2, EGLN3, EHD2 and VASN showed consistent associations with pubertal HOMA-IR z-score and, for several loci, with adiposity and blood pressure measures, with methylation changes paralleling IR worsening, improvement or stability across puberty. An mQTL look-up in GoDMC identified 25 cis SNP CpG associations corresponding to 20 of the 120 CpG sites, including CpG sites in SLC2A9 and TSC2, indicating that only a fraction of these IR-associated CpG sites is likely to be partly influenced by nearby genetic variants. Conclusion This longitudinal EWAS in children with obesity shows that specific blood DNA methylation signatures mirror IR status and track its evolution across the pubertal transition, with opposing methylation trajectories distinguishing improving from persistent IR. The identification of CpG sites at VASN, SLC2A9, PEPD, EGLN3, EHD2 and TSC2 links IR trajectories to pathways involved in vascular signalling, urate transport, extracellular matrix remodelling, and hypoxia sensing and nutrient signalling. Complementary mQTL analyses suggest that while some of this epigenetic variation is influenced by local genetic factors, a substantial component is likely acquired in response to metabolic and external exposures. If replicated and functionally characterised, these findings may help refine our understanding of the early molecular architecture of obesity-related IR and inform future strategies for cardiometabolic risk assessment and timing of preventive interventions.
Zaguán. Repositorio Digital de la Universidad de Zaragoza
Publication Artículo científico (JournalArticle)

Metabolomic profiling of ultra-processed food consumption and cardiovascular risk in Spanish preschool children: a cross-sectional study

  • Pérez-Acosta, Josué A.
  • García-Gavilán, Jesús F.
  • Margara-Escudero, Hernando J.
  • Aguilera, Concepción M.
  • Leis, Rosaura
  • Gheorghita, Irina
  • Miguel-Berges, María L.
  • Navas-Carretero, Santiago
  • Sorlí, José V.
  • Lázaro, Iolanda
  • Pastor-Villaescusa, Belén
  • Vázquez-Cobela, Rocío
  • Iglesias-Vázquez, Lucía
  • De Miguel-Etayo, Pilar
  • Vazquez-Bolea, Natalia
  • Jurado-Castro, José Manuel
  • Picáns-Leis, Rosaura
  • Escribano, Joaquín
  • Moreno, Luis A.
  • Torre-Aguilar, MJosé de la
  • Gil-Campos, Mercedes
  • Salas-Salvadó, Jordi
  • Babio, Nancy
Background Ultra-processed food (UPF) consumption has increased substantially among children worldwide and has been associated with early cardiovascular alterations. We aimed to identify plasma metabolomic profiles associated with UPF consumption in children aged 3–6 years and to assess their association with cardiovascular risk factors. Methods This cross-sectional analysis used baseline data from the Childhood Obesity Risk Assessment Longitudinal Study (CORALS), with children recruited between March 2019 and June 2022 through primary schools and health-care centres in seven Spanish cities. 1148 children were included (mean age 5.0 years [SD 1.1]; 50.6% girls) after exclusions due to medical conditions and incomplete or implausible dietary or metabolomic data. Energy-adjusted UPF intake (g/d) was assessed using a validated food frequency questionnaire and categorised with NOVA classification. 249 metabolites were quantified by NMR-based metabolomics, and lipid-related and amino acid-related profiles/scores were derived. A composite cardiovascular risk score (CVRs) was constructed by summing sex-specific, age-adjusted z-scores for fat-mass index, HDL-and LDL-cholesterol, triglycerides, mean arterial pressure, and Homoeostatic Model Assessment for Insulin Resistance. Findings 19 metabolites were associated with UPF consumption after false discovery rate correction (FDR). A lipid-related score was positively associated with CVRs (β = 0.511; 95% CI: 0.413–0.610; FDR <0.001) and the amino acid–related score was associated with a lower CVRs (β = −0.494; 95% CI: −0.667 to −0.321; FDR <0.001). Discrimination of cardiovascular risk beyond covariates alone was higher after inclusion of metabolomic scores, particularly when both scores were considered jointly (ΔAUC 0.092, 95% CI: 0.013–0.172; p = 0.020). Interpretation In Spanish pre-schoolers UPF-related metabolomic profiles were associated with different cardiovascular risk factors. These diet-related metabolomic scores capture biological correlates relevant to early cardiovascular health and align with early-life metabolic pattering, supporting future longitudinal studies and prevention strategies.