RESUMO
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.
RESUMO
Obesity in children is related to the development of cardiometabolic complications later in life, where molecular changes of visceral adipose tissue (VAT) and skeletal muscle tissue (SMT) have been proven to be fundamental. The aim of this study is to unveil the gene expression architecture of both tissues in a cohort of Spanish boys with obesity, using a clustering method known as weighted gene co-expression network analysis. For this purpose, we have followed a multi-objective analytic pipeline consisting of three main approaches; identification of gene co-expression clusters associated with childhood obesity, individually in VAT and SMT (intra-tissue, approach I); identification of gene co-expression clusters associated with obesity-metabolic alterations, individually in VAT and SMT (intra-tissue, approach II); and identification of gene co-expression clusters associated with obesity-metabolic alterations simultaneously in VAT and SMT (inter-tissue, approach III). In both tissues, we identified independent and inter-tissue gene co-expression signatures associated with obesity and cardiovascular risk, some of which exceeded multiple-test correction filters. In these signatures, we could identify some central hub genes (e.g., NDUFB8, GUCY1B1, KCNMA1, NPR2, PPP3CC) participating in relevant metabolic pathways exceeding multiple-testing correction filters. We identified the central hub genes PIK3R2, PPP3C and PTPN5 associated with MAPK signaling and insulin resistance terms. This is the first time that these genes have been associated with childhood obesity in both tissues. Therefore, they could be potential novel molecular targets for drugs and health interventions, opening new lines of research on the personalized care in this pathology. This work generates interesting hypotheses about the transcriptomics alterations underlying metabolic health alterations in obesity in the pediatric population.
Assuntos
Doenças Cardiovasculares , Obesidade Infantil , Masculino , Humanos , Criança , Transcriptoma/genética , Obesidade Infantil/genética , Obesidade Infantil/complicações , Obesidade Infantil/metabolismo , Perfilação da Expressão Gênica , Gordura Intra-Abdominal/metabolismo , Gordura Intra-Abdominal/patologia , Músculo Esquelético , Doenças Cardiovasculares/patologia , Proteínas Tirosina Fosfatases não Receptoras/metabolismoRESUMO
The use of machine learning techniques for the construction of predictive models of disease outcomes (based on omics and other types of molecular data) has gained enormous relevance in the last few years in the biomedical field. Nonetheless, the virtuosity of omics studies and machine learning tools are subject to the proper application of algorithms as well as the appropriate pre-processing and management of input omics and molecular data. Currently, many of the available approaches that use machine learning on omics data for predictive purposes make mistakes in several of the following key steps: experimental design, feature selection, data pre-processing, and algorithm selection. For this reason, we propose the current work as a guideline on how to confront the main challenges inherent to multi-omics human data. As such, a series of best practices and recommendations are also presented for each of the steps defined. In particular, the main particularities of each omics data layer, the most suitable preprocessing approaches for each source, and a compilation of best practices and tips for the study of disease development prediction using machine learning are described. Using examples of real data, we show how to address the key problems mentioned in multi-omics research (e.g., biological heterogeneity, technical noise, high dimensionality, presence of missing values, and class imbalance). Finally, we define the proposals for model improvement based on the results found, which serve as the bases for future work.
Assuntos
Obesidade Infantil , Criança , Humanos , Aprendizado de Máquina , AlgoritmosRESUMO
Extracellular matrix (ECM) remodeling plays important roles in both white adipose tissue (WAT) and the skeletal muscle (SM) metabolism. Excessive adipocyte hypertrophy causes fibrosis, inflammation, and metabolic dysfunction in adipose tissue, as well as impaired adipogenesis. Similarly, disturbed ECM remodeling in SM has metabolic consequences such as decreased insulin sensitivity. Most of described ECM molecular alterations have been associated with DNA sequence variation, alterations in gene expression patterns, and epigenetic modifications. Among others, the most important epigenetic mechanism by which cells are able to modulate their gene expression is DNA methylation. Epigenome-Wide Association Studies (EWAS) have become a powerful approach to identify DNA methylation variation associated with biological traits in humans. Likewise, Genome-Wide Association Studies (GWAS) and gene expression microarrays have allowed the study of whole-genome genetics and transcriptomics patterns in obesity and metabolic diseases. The aim of this review is to explore the molecular basis of ECM in WAT and SM remodeling in obesity and the consequences of metabolic complications. For that purpose, we reviewed scientific literature including all omics approaches reporting genetic, epigenetic, and transcriptomic (GWAS, EWAS, and RNA-seq or cDNA arrays) ECM-related alterations in WAT and SM as associated with metabolic dysfunction and obesity.