Leveraging polygenic functional enrichment to improve GWAS power

Abstract

Functional genomics data has the potential to increase GWAS power by identifying SNPs that have a higher prior probability of association. Here, we introduce a method that leverages polygenic functional enrichment to incorporate coding, conserved, regulatory and LD-related genomic annotations into association analyses. We show via simulations with real genotypes that the method, Functionally Informed Novel Discovery Of Risk loci (FINDOR), correctly controls the false-positive rate at null loci and attains a 9-38% increase in the number of independent associations detected at causal loci, depending on trait polygenicity and sample size. We applied FINDOR to 27 independent complex traits and diseases from the interim UK Biobank release (average N =130K). Averaged across traits, we attained a 13% increase in genome-wide significant loci detected (including a 20% increase for disease traits) compared to unweighted raw p-values that do not use functional data. We replicated the novel loci in independent UK Biobank and non-UK Biobank data, yielding a highly statistically significant replication slope (0.66-0.69) in each case. Finally, we applied FINDOR to the full UK Biobank release (average N =416K), attaining smaller relative improvements (consistent with simulations) but larger absolute improvements, detecting an additional 583 GWAS loci. In conclusion, leveraging functional enrichment using our method robustly increases GWAS power.

Genetic Dissection of Growth Traits in a Unique Chicken Advanced Intercross Line

“…The lead SNP at chr27:3,608,297 is associated with the shank length in F 9 AILs, which lies in the intron of the ZNF652 gene. Although the function of this gene has not been reported in detail, discoveries from human GWAS have replicated the significant correlation between ZNF652 and body height in two independent cohorts [rs35587648, p = 7 × 10 −42 in Lango Allen et al (2010) and rs2072153, p = 4 × 10 −8 in Kichaev et al (2019)]. Interestingly, three other genes at this locus, PHOSPHO1, IGF2BP1, and GIP, have been reported to be related to skeletal development.…”

Low-frequency variant functional architectures reveal strength of negative selection across coding and non-coding annotations

“…This underscores the advantages of large GWAS (with imputed genotypes obtained using large reference panels such as HRC 25 ), compared to WES or exome chip data, for querying low-frequency variation 14 . Furthermore, using functionally informed association tests that assign higher weight to low-frequency non-synonymous variants or CTS annotations should significantly improve power in these analyses 4,19,67 . Second, our results provide guidance for the design of association studies targeting rare (MAF < 0.5%) variants, which require large sequencing datasets 12 (Figure 4), coupled with our predictions of causal rare variant effect size variance (Figure 6d), suggest that in most instances these annotations do not harbor causal variants with large mean squared effect sizes (with brain-related annotations and traits as a notable exception; also see ref.…”

Accurate and Scalable Construction of Polygenic Scores in Large Biobank Data Sets

“…Besides genotype information, we also obtained traits. Specifically, following Márquez-Luna et al, 14 Privé et al, 25 Lloyd-Jones et al, 29 Kichaev et al, 66 25 we treated either self-reported or ICD10 cases as 1 and others as 0. For TA, we treated ''get very tanned'' as 1 and others as 0.…”

“…and the Neale lab (see Web Resources), we obtained 16 continuous traits that have an observed SNP heritability estimated to be above 0.1 and 9 binary traits that have a prevalence between 0.01 and 0.3. The 16 continuous traits include standing height (SH, n ¼ 335,473), platelet count (PLT, n ¼ 326,219), bone mineral density (BMD, n ¼ 193,397), basal metabolic rate (BMR, n ¼ 330,306), body mass index (BMI, n ¼ 335,106), red blood cell count (RBC, n ¼ 326,220), age at menarche (AM, n ¼ 180,061), RBC distribution width (RDW, n ¼ 326,218), eosinophils count (EOS, n ¼ 325,653), white blood cell count (WBC, n ¼ 326,216), forced vital capacity (FVC, n ¼ 306,637), forced expiratory volume (FEV1) versus FVC ratio (FFR, n ¼ 306,637), waist-hip ratio (WHR, n ¼ 335,568), neuroticism score (NS, n ¼ 273,107), systolic blood pressure (SBP, n ¼ 313,972), and years of education (YE, n ¼ 225,898) (see Neale lab in Web Resources) 14,29,66. The nine binary traits include prostate cancer (PRCA, n ¼ 147,408, prevalence ¼ 0.05), tanning ability (TA, n ¼ 329,458, prevalence ¼ 0.20), type II diabetes (T2D, n ¼ 329,355, prevalence ¼ 0.04), coronary artery disease (CAD, n ¼ 238,284, prevalence ¼ 0.05), rheumatoid arthritis (RA, n ¼ 232,309, prevalence ¼ 0.02), breast cancer (BRCA, n ¼ 170,148, prevalence ¼ 0.07), asthma (AS, n ¼ 306,381, prevalence ¼ 0.14), morning person (MP, n ¼ 300,143, prevalence ¼ 0.27), and depression (MDD, n ¼ 284,252, prevalence ¼ 0.08).…”