候选位点放大图

步骤 5 · 逐位点 · 图 11 张 · 发表版表格 0 个　（流程见 v5 定稿）

11 个候选各一张 locus zoom：暴露与结局在同一区域的信号形状，点色为与锚点变异的 r²。

图

Figure F5c. Regional association plot for the WARS locus on chromosome 14, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the WARS locus on chromosome 14, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.956 (lead variant 14:100376300, SNP.PP.H4 = 0.996).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 4195 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.996 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 60% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Method. coloc.abf (Giambartolomei et al. 2014, PLoS Genetics 10:e1004383), single causal variant per region, priors p1 = p2 = 1e-4 and p12 = 1e-5. The posterior probabilities shown here were recomputed for this figure from the same regional extracts used in the main analysis and agree with the frozen results to within 1e-6.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Complication endpoints from FinnGen R9 (general-population controls); type 1 diabetes GCST90824163; type 2 diabetes Mahajan et al. 2018 (non-UK Biobank). Exposures: cis-pQTLs for 1,954 plasma proteins from UK Biobank Pharma Proteomics Project (Olink Explore; discovery cohort, n = 34,557 European ancestry).

Data sources and sample sizes. Exposure: UKB-PPP plasma proteome (Olink Explore 3072), 34,557 European participants, 1,954 proteins with a cis-pQTL. Outcomes: FinnGen R9 -- Diabetic retinopathy 10,413 cases / 308,633 controls; Diabetic maculopathy 3,572 cases / 308,547 controls; Diabetic nephropathy 4,111 cases / 308,539 controls; Diabetic neuropathy 2,843 cases / 271,817 controls.

Figure F5c. Regional association plot for the IFNAR1 locus on chromosome 21, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the IFNAR1 locus on chromosome 21, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Maculopathy PP.H4 = 0.940 (lead variant 21:33353501, SNP.PP.H4 = 0.346).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 3854 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.346 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 61% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the NUDT5 locus on chromosome 10, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the NUDT5 locus on chromosome 10, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.977 (lead variant 10:12202080, SNP.PP.H4 = 0.992).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 5327 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.992 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 64% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the ERMAP locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the ERMAP locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Maculopathy PP.H4 = 0.967 (lead variant 1:42744270, SNP.PP.H4 = 0.654).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 3978 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.654 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 64% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the APOE locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the APOE locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Nephropathy PP.H4 = 0.998 (lead variant 19:44908684, SNP.PP.H4 = 1.000); Maculopathy PP.H4 = 0.917 (lead variant 19:44908684, SNP.PP.H4 = 1.000).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 4523 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 1.000 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 66% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the NOTCH2 locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the NOTCH2 locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Maculopathy PP.H4 = 0.897 (lead variant 1:119956951, SNP.PP.H4 = 0.061); Retinopathy PP.H4 = 0.896 (lead variant 1:119956951, SNP.PP.H4 = 0.060).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 2504 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.061 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 61% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the ACRBP locus on chromosome 12, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the ACRBP locus on chromosome 12, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.867 (lead variant 12:6645025, SNP.PP.H4 = 1.000).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 3832 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 1.000 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 45% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the GALNT3 locus on chromosome 2, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the GALNT3 locus on chromosome 2, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.948 (lead variant 2:165867023, SNP.PP.H4 = 0.566).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 4134 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.566 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 56% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the LACTB2 locus on chromosome 8, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the LACTB2 locus on chromosome 8, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.800 (lead variant 8:70894507, SNP.PP.H4 = 0.664).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 3702 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.664 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 0% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the PAM locus on chromosome 5, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the PAM locus on chromosome 5, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Retinopathy PP.H4 = 0.959 (lead variant 5:103011162, SNP.PP.H4 = 0.586).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 4086 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.586 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 52% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

Figure F5c. Regional association plot for the SIGLEC5 locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

完整图注（英文，与投稿版一致）

Figure F5c. Regional association plot for the SIGLEC5 locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below.

Colocalisation result. Maculopathy PP.H4 = 0.887 (lead variant 19:51627384, SNP.PP.H4 = 0.998).

What is plotted. Each point is one genetic variant. Only variants present in both the exposure and the outcome dataset are shown, that is, exactly the 5284 variants that entered the colocalisation analysis for this locus, within +/- 500 kb of the cis gene. P values are recovered from the effect estimate and its standard error because the regional extracts carry beta and standard error rather than P. The y axis is free between tracks: the two signals differ in magnitude by orders of magnitude and a shared axis would flatten the weaker one.

The black diamond and the dashed line mark the variant with the highest posterior probability of being the shared causal variant (SNP.PP.H4 = 0.998 from coloc.abf), not the instrument used in the Mendelian randomisation analysis. Points are coloured by linkage disequilibrium r-squared with that variant in the 1000 Genomes European reference panel; r-squared is available for 69% of variants and the remainder are grey. The reference panel is built on GRCh37 while the association data are GRCh38, so variants are matched by rsID and never by position.

Colocalisation is a wording gate in this study and does not remove candidates; loci that do not colocalise remain in all downstream tables.

发表版表格

本页的表格与共定位共用（同属 08_colocalisation）。

本页图为网页版 PNG（最宽 1600 px）。投稿用的矢量 PDF 与 600 dpi TIFF 体积较大，留在仓库 results/coloc/figures/，不随文档站分发。

候选位点放大图 ​

图 ​

Figure F5c. Regional association plot for the WARS locus on chromosome 14, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the IFNAR1 locus on chromosome 21, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the NUDT5 locus on chromosome 10, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the ERMAP locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the APOE locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the NOTCH2 locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the ACRBP locus on chromosome 12, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the GALNT3 locus on chromosome 2, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the LACTB2 locus on chromosome 8, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the PAM locus on chromosome 5, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

Figure F5c. Regional association plot for the SIGLEC5 locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below ​

发表版表格 ​

候选位点放大图

图

Figure F5c. Regional association plot for the WARS locus on chromosome 14, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the IFNAR1 locus on chromosome 21, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the NUDT5 locus on chromosome 10, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the ERMAP locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the APOE locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the NOTCH2 locus on chromosome 1, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the ACRBP locus on chromosome 12, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the GALNT3 locus on chromosome 2, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the LACTB2 locus on chromosome 8, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the PAM locus on chromosome 5, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

Figure F5c. Regional association plot for the SIGLEC5 locus on chromosome 19, showing the plasma cis-pQTL signal above and each colocalising complication GWAS below

发表版表格