Multivariate and gene-based association testing of sarcopenia: Bushehr Elderly Health Program (BEH)

Sarcopenia can be measured by a variety of indicators, and in many cases these indicators are quite positively correlated. With the aim of improving statistical model results, the objective of this study was to use multivariate methods to identify genetic variants affecting sarcopenia indices simultaneously.

GWAS analysis was performed based on data collected from 2426 Iranians aged 60 and over who were enrolled in the Bushehr Elderly Health program (BEH). DNA samples were collected from all subjects during this phase to measure prevalence of musculoskeletal disorders and risk factors. To analyze BEH DNA samples, we used a combination of Multiphen test, which is a linear combination of phenotypes most associated with genotypes, and GATES, a gene-based association test that can handle millions of SNP results efficiently and that can assess gene-level statistical significance.

The upper and lower 50 kb of the IL10 gene are extracted from chromosome 1 at the position (206940947, 206945839). The next step was done to calculate P-values for SNPs in this gene of SMI and handgrip using Multiphen (joint model). In the Gates method, these P-values are used to calculate the overall P-value (0.046). Given the fact that this value is less than 5%, it is clear that this gene has been effective in preventing sarcopenia in the Iranian elderly population. The tutorial describes how Multiphen and Gates can be used to analyze sarcopenia, a multifactorial disease. In this study, the gene Il10 (P-value = 0.046) was analyzed as a risk gene for sarcopenia.

GWAS (Genome-wide association study) is a primary method for identifying genetic variants that influence the phenotype of a disease. Multi-phenotype analysis, which evaluates multiple phenotypes associated with the disease, can, however, identify additional genetic associations associated with the disease. An alternative to univariate GWAS, Multiphen tests the most related linear combinations of multifactorial diseases using a multi-phenotypic approach rather than univariate GWAS. The overall P-value of the selected gene is determined by using Gates, a gene-based analysis.