Polygenicity and prevalence contribute to investigating the genetic signature of common diseases

Abstract

Abstract It was a great challenge in searching for the pathogenic genes of complicated disorders that are deleterious to human health. To search for characterized genetic signatures of common diseases, we developed a causal genotype combination pattern (CGCP) program previously. In this study, we used mathematics formula to infer the theoretical existence of linear relationship between variants frequency and prevalence, then observed the robustness with multiple computer simulations. By operating a whole-exome sequencing dataset in 781 psoriasis cases and 676 healthy controls of the Chinese Han population with a new strategy of the CGCP method, 620 genotype combinations were identified specific for psoriasis which covered range from 4.7% (37) to 10% (78) of all cases, and the frequency of each genotype was common (≥ 1%). All these genotypes merged into 134 genes, and 41 of them have been reported to be associated with psoriasis previously. Besides immune-related genes, we also identified 23 genes related to ATP metabolism. Benefit by the public data set of 1000 Genomes Project Phase III and the prevalence of psoriasis in many different ethnic populations got though literature search, a strong positive correlation and a linear regression model (y = 0.617*x + 4.79E-3) were established between the average frequency of these psoriasis-specific genotype combinations (x) and the prevalence of disease in different ethnicities (y), which may explain the differentiated prevalence in different populations. This strategy provides a new perspective to understand the characteristics of population genetics in common diseases.