Quantifying the genetic component of the metabolic syndrome using a novel proposal score and SNP-based heritability

Introduction. Metabolic syndrome (MetS) is a complex, multifactorial disease that poses a major public health problem. MetS increases the risk of coronary heart disease (CHD), atherosclerotic cardiovascular diseases (ASCVD), type 2 diabetes mellitus (T2DM), and all-cause mortality. Currently, there are a many different criteria that define MetS but the physiopathology is not completely understood both in terms of clinical progression and genetic contribution. Aims. The present work characterizes MetS components (obesity, hypertension, glucose, etc.) as one continuous phenotype and genetic components of the proposed MetS score were estimated using both family-based samples and population-based samples. Methods. In the first step, Confirmatory Factor Analysis (CFA) was used to select a model with the best fit. After the selection of the best factor structure and development an algorithm to calculate the score, heritability was performed in both pedigrees and SNPs/markers data. For the first sample, SOLAR (Sequential Oligogenic Linkage Analysis Routines) software was used to obtain the estimates. For the second sample, genetic variance components were calculated by fitting a linear mixed model (LMM) using two types of genetic relatedness matrices (Identity-By-Descend, IBD and Genome-Wide Complex Trait Analysis, GCTA), different levels of Linkage Disequilibrium (LD) pruning (0.20 – 0.80 and no LD pruning), and suggestive Genome-Wide Association Study (GWAS) SNPs. Results. According to the analyses, the best CFA model was the bifactor model; estimated coefficients were used to calculate the MetS score. The score showed good performance and good agreement compared to the International Diabetes Federation (IDF) criteria, the gold standard used for clinical diagnosis. With regards to the estimation of genetic variance, heritability was significant and ranged from 0.1 to 0.4 in whole samples and in all models. The heterogeneity of the results was due to the different samples and different types of matrix inputs into the LMMs. Heritability obtained using the GCTA matrix was significantly increased compared to the IBD matrix. No significant differences between family data and genetic data (markers) in Sardinia samples were observed using an LD threshold of 0.80 with no pruning. Conclusions. Evidence of complex interactions in metabolic syndrome and significant genetic contributions were obtained from these analyses. Increased knowledge of the environmental and genetic components could allow for better assessment and identification of patients with this syndrome.