University of Dundee
A comprehensive 1000 Genomes-based genome-wide association meta-analysis of
coronary artery disease
Nikpay, Majid; Goel, Anuj; Won, Hong-Hee; Hall, Leanne M.; Willenborg, Christina; Kanoni,
Stavroula
Published in:
Nature Genetics
DOI:
10.1038/ng.3396
Publication date:
2015
Document Version
Publisher's PDF, also known as Version of record
Link to publication in Discovery Research Portal
Citation for published version (APA):
Nikpay, M., Goel, A., Won, H-H., Hall, L. M., Willenborg, C., Kanoni, S., Saleheen, D., Kyriakou, T., Nelson, C.
P., Hopewell, J. C., Webb, T. R., Zeng, L., Dehghan, A., Elver, M., Armasu, S. M., Auro, K., Bjonnes, A.,
Chasman, D. I., Chen, S., ... Farrall, M. (2015). A comprehensive 1000 Genomes-based genome-wide
association meta-analysis of coronary artery disease. Nature Genetics, 47(10), 1121-1130.
https://doi.org/10.1038/ng.3396
General rights
Copyright and moral rights for the publications made accessible in Discovery Research Portal are retained by the authors and/or other
copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with
these rights.
• Users may download and print one copy of any publication from Discovery Research Portal for the purpose of private study or research.
• You may not further distribute the material or use it for any profit-making activity or commercial gain.
• You may freely distribute the URL identifying the publication in the public portal.
Take down policy
If you believe that this document breaches copyright please contact us providing details, and we will remove access to the work immediately
and investigate your claim.
Download date: 10. Aug. 2022
© 2015 Nature America, Inc. All rights reserved.
Nature GeNetics ADVANCE ONLINE PUBLICATION 1
A R T I C L E S
CAD is the main cause of death and disability worldwide and rep-
resents an archetypal common complex disease with both genetic
and environmental determinants
1,2
. Thus far, 48 genomic loci have
been found to harbor common SNPs in genome-wide significant
association with the disease. Previous GWAS of CAD have tested the
common disease–common variant hypothesis, with meta-analyses
typically based on HapMap imputation training sets or tagging SNP
arrays with up to 2.5 million SNPs (85% with MAF >0.05)
3,4
. The
1000 Genomes Project
5
has considerably expanded the coverage of
human genetic variation, especially for lower-frequency variants
and insertion-deletions (indels). We assembled 60,801 cases and
123,504 controls from 48 studies for a GWAS meta-analysis of CAD;
34,997 (57.5%) of the cases and 49,512 (40.1%) of the controls had been
previously included in our Metabochip-based CAD meta-analysis
(Supplementary Fig. 1) (ref. 3). Imputation was based on the 1000
Genomes Project phase 1 v3 training set with 38 million variants,
of which over half are low frequency (MAF < 0.005) and one-fifth
are common (MAF > 0.05). The majority (77%) of the participants
were of European ancestry; 13% and 6% were of South Asian (India
and Pakistan) and East Asian (China and Korea) ancestry, respec-
tively, with smaller samples of Hispanic and African Americans
(Supplementary Table 1). Case status was defined by an inclusive
CAD diagnosis (for example, myocardial infarction, acute coronary
syndrome, chronic stable angina or coronary stenosis of >50%).
After selecting variants that met the allele frequency (MAF > 0.005)
and imputation quality control criteria in at least 29 (>60%) of the
studies, 8.6 million SNPs and 836,000 (9%) indels were included in
the meta-analysis (Fig. 1); of these variants, 2.7 million (29%) were
low frequency (0.005 < MAF < 0.05).
RESULTS
Scanning for additive associations
The results of an additive genetic model meta-analysis are
summarized in Manhattan plots (Fig. 2 and Supplementary Fig. 2).
Existing knowledge of genetic variants affecting risk of coronary artery disease (CAD) is largely based on genome-wide
association study (GWAS) analysis of common SNPs. Leveraging phased haplotypes from the 1000 Genomes Project, we report
a GWAS meta-analysis of ~185,000 CAD cases and controls, interrogating 6.7 million common (minor allele frequency (MAF) >
0.05) and 2.7 million low-frequency (0.005 < MAF < 0.05) variants. In addition to confirming most known CAD-associated loci,
we identified ten new loci (eight additive and two recessive) that contain candidate casual genes newly implicating biological
processes in vessel walls. We observed intralocus allelic heterogeneity but little evidence of low-frequency variants with larger
effects and no evidence of synthetic association. Our analysis provides a comprehensive survey of the fine genetic architecture of
CAD, showing that genetic susceptibility to this common disease is largely determined by common SNPs of small effect size.
In total, 2,213 variants (7.6% indels) showed significant associa-
tions (P < 5 × 10
−8
) with CAD with a low false discovery rate (FDR
q value < 2.1 × 10
−4
). When these 2,213 variants were grouped into
loci, 8 represented regions not previously reported as being associ-
ated with CAD at genome-wide levels of significance (Fig. 2 and
Table 1). Of the 48 loci previously reported at genome-wide levels
of significance, 47 showed nominally significant associations
(Supplementary Table 2). The exception was rs6903956, the lead
SNP for the ADTRP-C6orf105 locus detected in Han Chinese
6
, which
previously showed no association in the Metabochip meta-analysis
of Europeans and South Asians
3
. Thirty-six previously reported loci
showed genome-wide significance (Supplementary Table 2). Monte
Carlo simulations, guided by published effect sizes, suggest that our
study was powered to detect 34 of the previously reported loci (95%
confidence interval (CI) = 31–41 loci) at genome-wide significance.
Hence, our findings are fully consistent with the previously identified
CAD-associated loci.
The majority of the loci showing GWAS significance in the
present analysis were well imputed (82% with imputation quality >0.9)
(Fig. 3a) and had small effect sizes (odds ratio (OR) < 1.25)
(Fig. 3b). An exception was the lead SNP in the newly associated
chromosome 7q36.1 (NOS3) locus, rs3918226, which was only
moderately well imputed (quality of 0.78), but the validity of this
association was supported by existing genotype data, as rs3918226
was present on the HumanCVD BeadChip for which data were
available for some of the cohorts used in the present analysis, thereby
allowing directly measured genotypes to be compared with imputed
genotypes (Supplementary Table 3) (ref. 7). Three additional lower-
frequency and moderately well-imputed SNPs in LPA and APOE
(Fig. 3a), which were not previously reported in CAD GWAS
3,4
, also
showed strong associations (LPA: rs10455872, P = 5.7 × 10
−39
and
rs3798220, P = 4.7 × 10
−9
; APOE: rs7412, P = 8.2 × 10
−11
). The LPA
SNPs have previously been shown to be strongly associated with CAD
in candidate gene studies based on experimental genotype data
7,8
.
A comprehensive 1000 Genomes–based genome-wide
association meta-analysis of coronary artery disease
A full list of authors and affiliations appears at the end of the paper.
Received 13 January; accepted 14 August; published online 7 September 2015; doi:10.1038/ng.3396
© 2015 Nature America, Inc. All rights reserved.
2 ADVANCE ONLINE PUBLICATION Nature GeNetics
A R T I C L E S
The minor allele of SNP rs7412 encodes the ε2 allele of APOE, and
it has been well documented that carriers of the ε2 allele have lower
cholesterol levels; significant protection from CAD by this allele
was confirmed in a large meta-analysis
9
and the Metabochip study
(P = 0.0009) (ref. 3). However, rs7412 is not present on most com-
mercially available genome-wide genotyping arrays and cannot be
imputed using HapMap reference panels, highlighting the value of
the expanded coverage of the 1000 Genomes Project reference panels.
Finally, SNP rs11591147 in PCSK9, which encodes the low-frequency
(MAF = 0.01) p.Arg46Leu substitution that has been associated with
low LDL (low-density lipoprotein) cholesterol levels and cardiopro-
tection
10–13
, was imperfectly imputed (imputation quality = 0.61).
Nonetheless, these data provide the strongest evidence yet for a
protective effect of this variant in CAD (P = 7.5 × 10
−6
).
Scanning for non-additive associations
Few GWAS of CAD have systematically scanned for associations
that include dominance effects, and few truly recessive loci have
been reported
14,15
. We used a recessive inheritance model to search
for susceptibility effects conferred by homozygosity for the minor
(less frequent) allele. Two new recessive susceptibility loci were iden-
tified with MAF = 0.09 and 0.36 and genotypic OR = 0.67 and 1.12,
respectively (Fig. 2 and Table 1); these loci showed very little evidence
of association under an additive model (Table 1). A supplementary
analysis applying a dominant model identified multiple strong asso-
ciations with variants, all of which overlapped with loci identified in
the analysis applying an additive model (Supplementary Table 4).
Myocardial infarction subphenotype analysis
Subgroup analysis in cases with a reported history of myocardial
infarction (~70% of the total number of cases) did not identify any
additional associations reaching genome-wide significance. The asso-
ciation results for the myocardial infarction subphenotype for the 48
previously known CAD-associated loci and the 8 new additive CAD-
associated loci discovered in this study are shown in Supplementary
Table 5. The odds ratios for the lead SNPs at 56 loci for the broader
CAD phenotype (full cohort) and the myocardial infarction subpheno-
type are compared in Supplementary Figure 3. Although, as expected,
the odds ratios were very similar for most of the loci, the odds ratios
for the ABO and HDAC9 loci were sufficiently distinct in the two
cohorts for their 95% confidence intervals to lie away from the line of
equality, suggesting that the ABO locus preferentially associates with
myocardial infarction and the HDAC9 locus preferentially associates
with stable coronary disease but not myocardial infarction per se.
FDR and heritability analysis
We performed a joint association analysis to search for evidence of
synthetic associations
16
, where multiple low-frequency susceptibility
variants at a locus might be in LD with a common variant discov-
ered as the lead variant in a GWAS, and to compile an FDR-defined
list of informative variants for annotation and heritability analysis
3
.
Variants that showed suggestive additive association (P < 5 × 10
−5
)
were assigned to 214 putative susceptibility loci of 2 cM centered
on each lead variant, and all variants in these loci were examined;
consequently, the search space for the joint analysis included 1,399,533
variants. Using GCTA software
17
to perform an approximate joint
association analysis (Online Methods), we identified 202 FDR
variants (q value < 0.05) in 129 loci (Supplementary Table 6) with
multiple (2–14) tightly linked variants, corresponding to 57% of the
putative CAD susceptibility loci. The 202 FDR variants were mostly
common (median MAF = 0.22) and well imputed (median impu-
tation quality = 0.97). Ninety-five variants (explaining 13.3 ± 0.4%
of CAD heritability) mapped to 44 significant loci from GWAS, and 93
variants (explaining 12.9 ± 0.4% of CAD heritability) mapped to loci
that included a previously reported significant variant from GWAS
analysis. One hundred nine variants (explaining a further 9.3 ± 0.3%
of CAD heritability) mapped to other loci. Fifteen low-frequency
(MAF < 0.05) variants explained only 2.1 ± 0.2% of CAD heritability,
indicating that our study was ~90% powered to detect OR >1.5 with
low-frequency variants (Supplementary Table 7).
Common variants showing typical GWAS signals might be
coupled with one or more low-frequency variants with relatively large
effects
16
. We found no evidence for such synthetic associations in
the joint association analysis; that is, all low-frequency variants were
either a lead variant or were jointly associated (q value < 0.05) with
a common variant. Twenty of the 202 FDR variants (9.9%) were
indels (4–14 bp in size) as compared to 8.8% of all the variants in
the meta-analysis (P = 0.60). Low-frequency variants (MAF < 0.05)
0.1
0.2
0.3
0.4
0.5
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
0.5
1.0
1.5
2.0
2.5
3.0
MAF
Median info
0.1
0.2
0.3
0.4
0.5
0.4
0.5
0.6
0.7
0.8
0.9
1.0
0
0.5
1.0
1.5
2.0
2.5
n (×10
5
)
n (×10
5
)
3.0
MAF
Median info
a
b
Figure 1 Comparing the 1000 Genomes Project and HapMap imputation
training sets. Spectra of MAFs and median imputation quality (median
info) scores showing the number (n) of variants in each bin. (a) The
distribution for the 9.4 million 1000 Genomes Project phase 1 v3
variants. (b) The distribution for 2.5 million HapMap 2 SNPs. Imputation
quality was calculated as the median of the respective values in up to 48
contributing studies; the imputation quality for genotyped variants was set
equal to 1.0. The 1000 Genomes Project training set includes more low-
frequency variants, many of which have imputation qualities >0.9.
© 2015 Nature America, Inc. All rights reserved.
Nature GeNetics ADVANCE ONLINE PUBLICATION 3
A R T I C L E S
were strikingly under-represented (6.9% ver-
sus 29.0%; P = 4.9 × 10
−12
), which may reflect
on the statistical power to detect the modest
effects associated with these variants.
Annotation and ENCODE analysis
Functional annotations were assigned to
the 9.4 million variants studied in the CAD
additive meta-analysis using ANNOVAR
software
18
(Supplementary Table 8). The
202 FDR variants were depleted in intergenic
regions (P = 2.5 × 10
−7
) and enriched in introns
(P = 0.00035). Variants were also assigned
to three sets of ENCODE (Encyclopedia of
DNA Elements) features, namely histone/chromatin modifications
(HMs), DNase I–hypersensitive sites (DHSs) and transcription factor
binding sites (TFBSs) (Supplementary Table 9). The FDR variants
showed independent enrichment across 11 cell types for the HM
(P = 2.8 × 10
−6
) and DHS (P = 0.0003) ENCODE feature sets and
with genic annotation status (P = 0.0013) (Supplementary Tables 10
and 11). These associations were also evident in three cell types
selected for maximal CAD relevance, with a 2.6-fold enrichment
for DHSs, a 2.2-fold enrichment for HMs and a 1.6-fold enrichment
for genic status (Supplementary Tables 12 and 13). These findings
suggest that the 202 FDR variants are enriched for functional variants
with potential relevance to CAD pathogenesis.
Post -hoc power calculations
Of the 9.4 million variants analyzed, 8.2 million (87%) were highly
powered (>90%) to detect an OR ≥1.3 (Supplementary Table 7). The
number of variants with power of ≥90% to detect associations varied
systematically with allele frequency and imputation quality (results
for OR = 1.3 shown in Supplementary Fig. 4); 1.5 million of the
2.7 million (55%) low-frequency variants (0.005 < MAF < 0.05) in the
meta-analysis were adequately powered to detect an OR ≥1.3, as most of
these variants were accurately imputed (median imputation quality = 0.94,
interquartile range = 0.88–0.98). Of the more common variants (MAF
> 0.05), almost all (99.8%) were highly powered to detect an OR ≥1.3.
However, in terms of total coverage of low-frequency variation, only
15.3% of the 9.3 million low-frequency variants (0.005 < MAF < 0.05)
in the 1000 Genomes Project phase 1 v3 training set met the
allele frequency and imputation quality entry criteria in the 60% of the
studies required for inclusion in the meta-analysis and were predicted
to be adequately powered to detect significant associations; 100% of
these variants were highly powered (>90%) to detect an OR ≥3.15.
Interrogation of ten newly identified additive and recessive loci
We examined whether there were any expression quantitative trait loci
(eQTLs), associations with known cardiovascular risk factors or prior
evidence of the involvement of genes with atherosclerotic processes
in each of the newly identified loci to define putative mechanisms by
which the loci might affect risk of CAD.
At the chromosome 4q12 (REST-NOA1) locus, the lead SNP
rs17087335 lies within an intron of the NOA1 gene (nitric oxide–
associated 1); 23 SNPs in LD (r
2
> 0.8) showed CAD associations
(P < 1 × 10
−6
) across the NOA1 and REST (repressor element-1
silencing transcription factor) genes (Fig. 4a). NOA1 encodes a
GTP-binding protein involved in the regulation of mitochondrial
respiration and apoptosis
19
. REST encodes a transcription factor that
suppresses the expression of voltage-dependent sodium and potassium
channels
20
; it has been shown to maintain vascular smooth muscle
cells (VSMCs) in a quiescent, non-proliferative state and is itself
downregulated in neointimal hyperplasia
21
. SNP rs17087335 showed
a cis-eQTL signal for REST in lung
22
(Supplementary Table 14).
At the chromosome 7q36.1 (NOS3) locus, the lead SNP rs3918226
(MAF = 0.07) lies in the first intron of NOS3 (nitric oxide synthase 3)
(Fig. 4b). This SNP was tentatively associated with CAD (OR = 1.14,
P = 1.4 × 10
−4
) in a candidate gene meta-analysis based on 15,600
Figure 2 A circular Manhattan plot
summarizing the 1000 Genomes Project
CAD association results. The meta-analysis
statistics were adjusted for overdispersion
(before applying double genomic control,
λ
= 1.18); overdispersion is predicted to be
a regular feature in GWAS under a polygenic
inheritance model
60
. The association statistics
were capped at P = 1 × 10
−20
. Genome-wide
significant variants (P < 5 × 10
−8
) are indicated
by red triangles. New CAD-associated loci are
indicated by red text (Table 1). Previously
reported loci showing genome-wide significant
association are indicated by black text, and
those showing nominal significance (P < 0.05)
in our meta-analysis are indicated by blue
text (Supplementary Table 2). The inner track
shows the imputation quality scores of the
lead variants in the new loci. The middle track
shows numbered chromosome ideograms with
centromeres represented by pink bars.
© 2015 Nature America, Inc. All rights reserved.
4 ADVANCE ONLINE PUBLICATION Nature GeNetics
A R T I C L E S
cases and 35,000 controls genotyped with the HumanCVD BeadChip
7
and was firmly associated with essential hypertension (OR = 1.34,
P = 1.0 × 10
−14
) (ref. 23). NOS3 is involved in the production of nitric
oxide (NO), a potent vascular smooth muscle relaxant, and is a well-
studied candidate gene for CAD. Indeed, the genes encoding the com-
ponents of the NO receptor (soluble guanylyl cyclase) display both
linkage and genome-wide association with CAD
3,24
. There are several
overlapping ENCODE features in intron 1 of NOS3, suggesting a func-
tional role for rs3918226. However, there are 30 genes neighboring
NOS3 within a 2-cM window centered on this variant, and the cur-
rent data do not allow the candidacy of these genes to be excluded. A
nonsynonymous SNP, rs1799983, in NOS3 previously associated with
cardiovascular phenotypes
25
is in weak LD with rs3918226 but did not
achieve significance in the additive or joint association analysis.
At the chromosome 11p15.4 (SWAP70) locus, SNP rs10840293 is
intronic to SWAP70 (switch-associated protein-70) (Fig. 4c). SWAP-
70 is a signaling molecule involved in the regulation of filamentous
actin networks
26
in cell migration and adhesion. SNP rs10840293 and
other SNPs in strong LD are cis eQTLs for SWAP70 in naive and chal-
lenged monocytes
27
, with SNP rs93138 showing strong association
with CAD (P = 5.5 × 10
−8
) and being a cis eQTL for SWAP70 in naive
and challenged monocytes
28
, fat
29
, skin
29
and lung
22
(Supplementary
Table 14); three of the linked SNPs (rs93138, rs173396 and rs472109)
are intronic and lie within ENCODE regulatory functional elements.
Although this CAD-associated locus includes 33 genes, the eQTL
and ENCODE data implicate SWAP70 as a plausible causal gene and
suggest putative causal SNPs.
At the chromosome 15q22.33 (SMAD3) locus, the lead SNP
rs56062135 is intronic to SMAD3 and the CAD association is tightly
localized between two recombination hot spots (Fig. 4d). Mice lacking
Smad3, a major downstream mediator of transforming growth
factor (TGF)-β signaling, show enhanced neointimal hyperplasia
with decreased matrix deposition in response to vascular injury
30
.
SMAD3 was tentatively associated with CAD in an earlier GWAS
31
,
although the lead SNP (rs17228212) in that association is in linkage
rs180803
rs3918226
rs8042271
rs10455872
rs3798220
rs7412
rs12976411
rs11830157
1.00
Imputation quality
0 0.1 0.2 0.3 0.4 0.5
MAF
rs2891168
rs55730499
rs180803
rs12976411
rs11830157
Odds ratio
0 0.2 0.4 0.6 0.8 1.0
0.95
0.85
0.80
0.90
1.5
1.4
1.3
1.2
1.1
1.0
EAF
a
b
Figure 3 The imputation quality and effect size of lead variants at
46 genome-wide significant loci. (a) The imputation quality and MAF
for the lead variants at 46 genome-wide significant susceptibility loci.
Blue circles, new additive loci; red squares, new recessive loci; black
triangles, previously mapped additive loci; black diamonds, key SNPs
in LPA and APOE. Imputation quality and MAF were each calculated as
the median of the respective values in up to 48 contributing studies;
the imputation quality for studies with genotype data was fixed at 1.0.
(b) The odds ratio and effect allele frequency (EAF) for the lead variants
at 46 genome-wide significant loci. Blue circles, new additive loci; red
squares, new recessive loci; black triangles, previously mapped additive
loci. SNPs rs55730499 and rs2891168 are lead variants in the LPA
and chromosome 9p21 susceptibility loci, respectively. EAF was
calculated as the median value in up to 48 contributing studies.
Table 1 Ten new CAD-associated loci
Lead variant Locus name Chr. A1/A2
Effect
allele
(A1)
freq.
Imputation
quality I
2
Heterogeneity
P
n
studies
a
Association model
Additive Recessive
OR (95% CI) P OR (95% CI) P
rs17087335 REST-NOA1 4 T/G 0.21 0.99 0.20 0.11 48 1.06 (1.04–1.09) 4.60 × 10
−8
1.11 (1.05–1.17) 3.30 × 10
−4
rs3918226 NOS3 7 T/C 0.06 0.78 0.15 0.19 45 1.14 (1.09–1.19) 1.70 × 10
−9
1.26 (0.99–1.60) 5.96 × 10
−2
rs10840293 SWAP70 11 A/G 0.55 0.94 0.17 0.16 47 1.06 (1.04–1.08) 1.30 × 10
−8
1.05 (1.02–1.09) 1.51 × 10
−3
rs56062135 SMAD3 15 C/T 0.79 0.98 0.00 0.67 46 1.07 (1.05–1.10) 4.50 × 10
−9
1.17 (1.10–1.25) 8.88 × 10
−7
rs8042271 MFGE8-ABHD2 15 G/A 0.9 0.93 0.16 0.19 46 1.10 (1.06–1.14) 3.70 × 10
−8
1.25 (1.13–1.37) 7.27 × 10
−6
rs7212798 BCAS3 17 C/T 0.15 0.95 0.14 0.21 48 1.08 (1.05–1.11) 1.90 × 10
−8
1.17 (1.07–1.28) 6.12 × 10
−4
rs663129 PMAIP1-MC4R 18 A/G 0.26 1.00 0.00 0.6 47 1.06 (1.04–1.08) 3.20 × 10
−8
1.11 (1.06–1.17) 7.15 × 10
−6
rs180803 POM121L9P-
ADORA2A
22 G/T 0.97 0.86 0.00 0.67 41 1.20 (1.13–1.27) 1.60 × 10
−10
NA NA
rs11830157 KSR2 12 G/T 0.36 0.99 0.14 0.22 42 1.04 (1.02–1.06) 3.90 × 10
−4
1.12 (1.08–1.16) 2.12 × 10
−9
rs12976411 ZNF507-
LOC400684
19 T/A 0.09 0.93 0.50 5.09 × 10
−4
34 0.95 (0.92–0.99) 9.10 × 10
−3
0.67 (0.60–0.74) 1.18 × 10
−14
Association results are presented for two inheritance models; results from the discovery association model are shown in bold. P values were adjusted for overdispersion
following meta-analysis. Heterogeneity P values are for the respective discovery association model. Chr., chromosome; A1, effect allele; A2, non-effect allele; freq., frequency;
I
2
, heterogeneity inconsistency index; OR, odds ratio; CI, confidence interval; NA, not available owing to insufficient numbers (<60%) of studies having reliable results.
a
The number of studies that participated in the discovery result, where up to 48 studies participated in the additive model meta-analysis and up to 43 studies participated in the recessive
model meta-analysis.