Quantitative Research Of Missing Heritability In Genome Wide Association Study Of Psoriasis

Background: Psoriasis is a common immune-mediated skin disease characterized by epidermal hyperplasia, inflammatory cell infiltration, and vascular remodeling. Now it is well considered as a complex genetic disease resulted from the interactions between genetic variants and environmental factors. However, the exact etiology is largely unclear. Recently, benefited from the complete of International Hap Map Project and with the advent of high throughput genotyping array, genome-wide association studies(GWASs) have been implemented in the identification of susceptibility genes for complex diseases or traits, now it becomes one of the most popular methods in this context. Several international research groups have developed GWASs for psoriasis susceptibility gene in different races in large sample cases, and have identified more than 40 susceptibility genes and effectively revealed the genetic etiology of psoriasis. However, these previously identified susceptibilities only explained collectively a small fraction of estimated heritability of psoriasis, which was termed as “Missing Heritability” in GWASs. To integrate and excavate GWAS data, and revealing the “Missing Heritability” in psoriasis GWAS and exploring the possible existence of unknown susceptibility genesis become the most popular and hardest task in the perspective of psoriasis susceptibility.Objectives: To integrate and excavate GWAS data, and characterize the polygenic architecture and reveal the “Missing Heritability”, and provide the direction for the follow-up study of psoriasis susceptibility gene.Methods: We implemented the phenotype correlation-genetic correlation(PCGC) regression in the genome-wide association dataset(1139 PV patients and 1132 healthy control) to characterize the polygenic architecture and quantify the missingheritability in psoriasis GWAS in Han Chinese population, based on the early stage of mixed linear model study. We performed a genotype–phenotype relationship analysis to explore the relationships of the susceptibility gene FLT3(rs34172843) with Ps V subphenotypes in a Chinese Han population, which was identified in our exome array study, to derive the unknown susceptibility gene and provide the direction for the follow-up study of psoriasis susceptibility gene.Results:1. We performed the phenotype correlation-genetic correlation(PCGC) regression using genome-wide association data in 2,271 Han Chinese individuals. We estimated that 40.6%(s.e.=11.0%) of the variation in the liability to psoriasis is captured by common variants. Through genotyped and imputed, we found 45.7%(s.e.=12.5%) of the variation in the liability to psoriasis is captured. By Analyzing the genotyped and imputed SNPs of HLA separately, we can explain the 11.7%(s.e.=2.1%) of the variation in the liability to psoriasis. We extracted SNPs from 12 previously identified susceptibility regions in our previous GWAS, estimated h2 SNP to be 16.2%(s.e.=3.4%), We also show that the variance explained by each chromosome is linearly correlated to its length(R2=0.87, P=6.52×10-10). Rare variant can only cover 0.8%(s.e.=6.9%) in the liability to psoriasis,while covering 30.2%(s.e.=7.2%) in common variant. 2. Genotype- phenotype analysis: In this study, allelic frequency distribution analysis of rs34172843 of FLT3 was significant difference between cases and controls(P= 5.2×10-3, OR=1.15,95%CI=1.04-1.27). 2.1 Onset-age: Allele frequency distribution analysis of rs34172843 was no significant difference between later onset and early onset(P=4.64×10-1, OR=1.08, 95%CI=0.88-1.32). Allele frequency distribution analysis of rs34172843 was no significant difference between later onset and controls(P=4.53×10-1,OR=1.08,95%CI=0.89-1.32), but there was significant difference between the early onset and controls(P=4.06×10-3, OR=1.16, 95%CI=1.05-1.29). 2.2 Severity: Allele frequency distribution analysis of rs34172843 was significantdifference between the mild and moderate(P=1.44×10-2,OR=1.20,95%CI=1.04-1.41),and was no significant difference between the moderate and severe(P=7.81×10-1, OR=0.96, 95%CI=0.70-1.31),and was also no significant difference between the mild and severe(P =3.39×10-1, OR=0.96,95%CI=0.70-1.31). Allele frequency distribution analysis of rs34172843 was significant difference between the mild and controls(P=2.14×10-4,OR=1.24, 95%CI=1.11-1.38).But, allele frequency distribution analysis of rs34172843 was no significant difference between the moderate and controls(P=7.56×10-1,OR=1.02,95%CI=0.89-1.18),while there was no significant difference between the severe and controls(P=6.50×10-1, OR=1.07, 95%CI=0.80-1.44). 2.3 Skin lesion: Allele frequency distribution analysis of rs34172843 was significant difference between the guttate and plaque(P=4.76×10-2, OR=1.16,95%CI=1.00-1.35). Allele frequency distribution analysis of rs34172843 was significant difference between the guttate and controls(P=7.13×10-4, OR=1.28, 95%CI=1.11-1.47),but there was no significant difference between the plaque and controls(P=9.34×10-2,OR=1.10, 95%CI=0.98-1.23). 2.4 Family history: Allele frequency distribution analysis of rs34172843 was no significant difference between family history cases and no family history cases(P=9.06×10-1, OR=1.01, 95%CI=0.84-1.22). Allele frequency distribution analysis of rs34172843 was no significant difference between family history cases and controls(P=1.13×10-1, OR=1.16, 95%CI=0.97-1.40), but there was significant difference between no family history cases and controls(P=9.10×10-3, OR=1.15, 95%CI=1.04-1.27).Conclusions: This study has first performed phenotype correlation-genetic correlation(PCGC) regression in the genome-wide association dataset, carried out a genotype–phenotype relationship analysis for association site of no reaching standard threshold of GWAS(P>5×10-8). We fully explore the polygenic architecture, and confirm the presence of "missing heritability" in genome-wide association analysis inpsoriasis, and quantify the "missing heritability". By integrating and excavating genome-wide association data, we can further explore the genetic etiology of psoriasis, and explore pathogenesis of psoriasis, and make individualized diagnosis and treatment, and develop the targeted drugs.