4  SOST region associations

Genetic associations in the SOST region for the datasets in Table 2.2. 1

# Data
load("./data/01_data_ldmat.Rda")
load("./data/02_data_gwas_sost_region.Rda")

# SOST region
gwas <- gwas %>%
  filter(
    chr == "17" &
      pos >= 41831099 - 100000 &
      pos <= 41836156 + 100000
  )

4.1 Original datasets

Genetic associations in the SOST region for the datasets in Table 2.2 that were analysed by Zheng et al. (2023) (or equivalent2).

gwas %>%
  filter(id == "GCST011365") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.1: Regional plot of the SOST region for myocardial infarction (GCST011365)
gwas %>%
  filter(id == "GCST011365") %>%
  qq_plot()
Figure 4.2: QQ plot of the SOST region for myocardial infarction (GCST011365)
gwas %>%
  filter(id == "GCST005194") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.3: Regional plot of the SOST region for coronary artery disease (GCST005194)
gwas %>%
  filter(id == "GCST005194") %>%
  qq_plot()
Figure 4.4: QQ plot of the SOST region for coronary artery disease (GCST005194)
gwas %>%
  filter(id == "GCST005843") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.5: Regional plot of the SOST region for ischemic stroke (GCST005843)
gwas %>%
  filter(id == "GCST005843") %>%
  qq_plot()
Figure 4.6: QQ plot of the SOST region for ischemic stroke (GCST005843)
gwas %>%
  filter(id == "GCST005842") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.7: Regional plot of the SOST region for cardioembolic stroke (GCST005842)
gwas %>%
  filter(id == "GCST005842") %>%
  qq_plot()
Figure 4.8: QQ plot of the SOST region for cardioembolic stroke (GCST005842)
gwas %>%
  filter(id == "GCST006867") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs1107747",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.9: Regional plot of the SOST region for type 2 diabetes (GCST006867)
gwas %>%
  filter(id == "GCST006867") %>%
  qq_plot()
Figure 4.10: QQ plot of the SOST region for type 2 diabetes (GCST006867)
gwas %>%
  filter(id == "UKB200021065") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.11: Regional plot of the SOST region for hypertension (UKB200021065)
gwas %>%
  filter(id == "UKB200021065") %>%
  qq_plot()
Figure 4.12: QQ plot of the SOST region for hypertension (UKB200021065)
gwas %>%
  filter(id == "PHS000930AAC") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.13: Regional plot of the SOST region for abdominal aortic calcification (PHS000930AAC)
gwas %>%
  filter(id == "PHS000930AAC") %>%
  qq_plot()
Figure 4.14: QQ plot of the SOST region for abdominal aortic calcification (PHS000930AAC)
gwas %>%
  filter(id == "UKB30780") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.15: Regional plot of the SOST region for LDL cholesterol (UKB30780)
gwas %>%
  filter(id == "UKB30780") %>%
  qq_plot()
Figure 4.16: QQ plot of the SOST region for LDL cholesterol (UKB30780)
gwas %>%
  filter(id == "UKB30760") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants",
    highlights_label = FALSE
  )
Figure 4.17: Regional plot of the SOST region for HDL cholesterol (UKB30760)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30760") %>%
  qq_plot()
Figure 4.18: QQ plot of the SOST region for HDL cholesterol (UKB30760)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30870") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants",
    highlights_label = FALSE
  )
Figure 4.19: Regional plot of the SOST region for triglycerides (UKB30870)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30870") %>%
  qq_plot()
Figure 4.20: QQ plot of the SOST region for triglycerides (UKB30870)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30630") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants",
    highlights_label = FALSE
  )
Figure 4.21: Regional plot of the SOST region for apolipoprotein A-I (UKB30630)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30630") %>%
  qq_plot()
Figure 4.22: QQ plot of the SOST region for apolipoprotein A-I (UKB30630)

Note: rs72836561 is a missense variant in CD300LG that is known to be associated with lipid profiles (Surakka et al. 2015).

gwas %>%
  filter(id == "UKB30640") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.23: Regional plot of the SOST region for apolipoprotein B (UKB30640)
gwas %>%
  filter(id == "UKB30640") %>%
  qq_plot()
Figure 4.24: QQ plot of the SOST region for apolipoprotein B (UKB30640)
gwas %>%
  filter(id == "GCST006980") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.25: Regional plot of the SOST region for fracture (GCST006980)
gwas %>%
  filter(id == "GCST006980") %>%
  qq_plot()
Figure 4.26: QQ plot of the SOST region for fracture (GCST006980)
gwas %>%
  filter(id == "GCST006979") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.27: Regional plot of the SOST region for heel bone mineral density (GCST006979)
gwas %>%
  filter(id == "GCST006979") %>%
  qq_plot()
Figure 4.28: QQ plot of the SOST region for heel bone mineral density (GCST006979)

4.2 Additional datasets

Genetic associations in the SOST region for the additional recent GWAS datasets in Table 2.2.

gwas %>%
  filter(id == "GCST90132315") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.29: Regional plot of the SOST region for coronary artery disease (GCST90132315)

Note: the allele of rs851058 associated with lower coronary artery disease risk is also associated with lower mRNA expression of the SOST in GTEx (GTEx Consortium 2020) and higher bone mineral density (Morris et al. 2019).

gwas %>%
  filter(id == "GCST90132315") %>%
  qq_plot()
Figure 4.30: QQ plot of the SOST region for coronary artery disease (GCST90132315)

Note: the allele of rs851058 associated with lower coronary artery disease risk is also associated with lower mRNA expression of the SOST in GTEx (GTEx Consortium 2020) and higher bone mineral density (Morris et al. 2019).

gwas %>%
  filter(id == "GCST90104535") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.31: Regional plot of the SOST region for ischemic stroke (GCST90104535)
gwas %>%
  filter(id == "GCST90104535") %>%
  qq_plot()
Figure 4.32: QQ plot of the SOST region for ischemic stroke (GCST90104535)
gwas %>%
  filter(id == "GCST90104536") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.33: Regional plot of the SOST region for cardioembolic stroke (GCST90104536)
gwas %>%
  filter(id == "GCST90104536") %>%
  qq_plot()
Figure 4.34: QQ plot of the SOST region for cardioembolic stroke (GCST90104536)
gwas %>%
  filter(id == "GCST90104538") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.35: Regional plot of the SOST region for large artery stroke (GCST90104538)
gwas %>%
  filter(id == "GCST90104538") %>%
  qq_plot()
Figure 4.36: QQ plot of the SOST region for large artery stroke (GCST90104538)
gwas %>%
  filter(id == "GCST90104537") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.37: Regional plot of the SOST region for small vessel stroke (GCST90104537)
gwas %>%
  filter(id == "GCST90104537") %>%
  qq_plot()
Figure 4.38: QQ plot of the SOST region for small vessel stroke (GCST90104537)
gwas %>%
  filter(id == "GCST90132183") %>%
  regional_plot(
    data = .,
    corr = ld_mat,
    top_marker = "rs66838809",
    highlights = pqtls$rsid,
    highlights_title = "Sclerostin associated variants"
  )
Figure 4.39: Regional plot of the SOST region for type 2 diabetes (GCST90132183)
gwas %>%
  filter(id == "GCST90132183") %>%
  qq_plot()
Figure 4.40: QQ plot of the SOST region for type 2 diabetes (GCST90132183)

  1. The red and green dashed lines in the regional plots are the p-value thresholds \(1 \times 10^{-6}\) (regional-wide significance threshold used in Zheng et al. (2023)) and \(5 \times 10^{-8}\) (genome-wide significance threshold), respectively.↩︎

  2. The ischemic and cardioembolic stroke GWAS results from METASTROKE (Malik et al. 2016) used by Zheng et al. (2023) were replaced with those from MEGASTROKE (Malik et al. 2018) and the UK Biobank hypertension GWAS results from OpenGWAS used by Zheng et al. (2023) were replaced with those from Pan-UKBB due to licensing restrictions. The GWAS of coronary artery calcification was not available either publicly or via application at the time of this analysis (Kavousi et al. 2022).↩︎