VU Research Portal
Genome-wide association analyses identify 44 risk variants and refine the genetic
architecture of major depression
Wray, Naomi R.; Ripke, Stephan; Mattheisen, Manuel; Trzaskowski, MacIej; Byrne, Enda
M.; Abdellaoui, Abdel; Adams, Mark J.; Agerbo, Esben; Air, Tracy M.; Andlauer, Till M.F.;
Bacanu, Silviu Alin; Bækvad-Hansen, Marie; Beekman, Aartjan F.T.; Bigdeli, Tim B.;
Binder, Elisabeth B.; Blackwood, Douglas R.H.; Bryois, Julien; Buttenschøn, Henriette N.;
Bybjerg-Grauholm, Jonas; Cai, Na
published in
Nature Genetics
2018
DOI (link to publisher)
10.1038/s41588-018-0090-3
document version
Publisher's PDF, also known as Version of record
document license
Article 25fa Dutch Copyright Act
Link to publication in VU Research Portal
citation for published version (APA)
Wray, N. R., Ripke, S., Mattheisen, M., Trzaskowski, M., Byrne, E. M., Abdellaoui, A., Adams, M. J., Agerbo, E.,
Air, T. M., Andlauer, T. M. F., Bacanu, S. A., Bækvad-Hansen, M., Beekman, A. F. T., Bigdeli, T. B., Binder, E.
B., Blackwood, D. R. H., Bryois, J., Buttenschøn, H. N., Bybjerg-Grauholm, J., ... Penninx, B. W. J. H. (2018).
Genome-wide association analyses identify 44 risk variants and refine the genetic architecture of major
depression. Nature Genetics, 50(5), 668-681. https://doi.org/10.1038/s41588-018-0090-3
General rights
Copyright and moral rights for the publications made accessible in the public 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 the public 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.
E-mail address:
vuresearchportal.ub@vu.nl
Download date: 09. Aug. 2022
Articles
https://doi.org/10.1038/s41588-018-0090-3
Major depressive disorder (MDD) is a common illness accompanied by considerable morbidity, mortality, costs, and heightened
risk of suicide. We conducted a genome-wide association meta-analysis based in 135,458 cases and 344,901 controls and iden-
tified 44 independent and significant loci. The genetic findings were associated with clinical features of major depression and
implicated brain regions exhibiting anatomical differences in cases. Targets of antidepressant medications and genes involved
in gene splicing were enriched for smaller association signal. We found important relationships of genetic risk for major depres-
sion with educational attainment, body mass, and schizophrenia: lower educational attainment and higher body mass were
putatively causal, whereas major depression and schizophrenia reflected a partly shared biological etiology. All humans carry
lesser or greater numbers of genetic risk factors for major depression. These findings help refine the basis of major depression
and imply that a continuous measure of risk underlies the clinical phenotype.
M
DD is a notably complex and common illness
1
. It is often
chronic or recurrent and is thus accompanied by consider
-
able morbidity, disability, excess mortality, substantial costs,
and heightened risk of suicide
2–8
. Twin studies attribute approximately
40% of the variation in liability to MDD to additive genetic effects
(phenotype heritability, h
2
)
9
, and h
2
may be greater for recurrent,
early-onset, and postpartum MDD
10,11
. Genome-wide association
studies (GWAS) of MDD have had notable difficulties in identifying
individual associated loci
12
. For example, there were no significant
findings in the initial Psychiatric Genomics Consortium (PGC) MDD
mega-analysis (9,240 cases)
13
or in the CHARGE meta-analysis of
depressive symptoms (n = 34,549)
14
. More recent studies have proven
modestly successful. A study of Han Chinese women (5,303 recurrent
MDD cases) identified significant loci
15
, a meta-analysis of depressive
symptoms (161,460 individuals) identified 2 loci
16
, and an analysis of
self-reported major depression identified 15 loci (75,607 cases).
There are many reasons why identifying causal loci for MDD has
proven difficult
12
. MDD is probably influenced by many genetic loci,
each with small effects
17
, as are most common diseases
18
, including psy-
chiatric disorders
19,20
. Estimates of the proportion of variance attribut-
able to genome-wide SNPs (SNP heritability,
h
SNP
2
) indicate that around
one-quarter of the h
2
for MDD is due to common genetic variants
21,22
and demonstrate that a genetic signal is detectable in genome-wide
association data, implying that larger sample sizes are needed to detect
specific loci given their effect sizes. Such a strategy has been proven
in studies of schizophrenia, the flagship adult psychiatric disorder in
genomics research. We thus accumulated clinical, population, and vol
-
unteer cohorts
23
. This pragmatic approach takes the view that sample
size can overcome heterogeneity to identify risk alleles that are robustly
associated with major depression. Potential concerns about combining
carefully curated research cohorts with volunteer cohorts were consid
-
ered; multiple lines of evidence suggest that the results are likely to be
applicable to clinical MDD. As discussed below, our analyses have neu
-
robiological, clinical, and therapeutic relevance for major depression.
Results
Cohort analyses: phenotype validation. We identified seven
cohorts that used a range of methods to ascertain cases with major
depression (Table1 and Supplementary Tables1–3). The methods
used by these cohorts were thoroughly reviewed, drawing on the
breadth of expertise in the PGC, and we assessed the comparability
of the cohorts using genomic data. We use ‘MDD’ to refer to directly
evaluated subjects meeting standard criteria for major depressive
disorder and use ‘major depression’ where case status was deter
-
mined using alternative methods as well as to the phenotype from
the full meta-analysis.
We evaluated the comparability of the seven cohorts by esti
-
mating the common variant genetic correlations (r
g
) between
them. These analyses supported the comparability of the seven
cohorts (Supplementary Table3), as the weighted mean r
g
was 0.76
(s.e. = 0.03). The high genetic correlations between the 23andMe
and other cohorts are notable. While there was no statistical evi
-
dence of heterogeneity in the r
g
estimates for pairs of cohorts
(P = 0.13), the estimate was statistically different from 1, which may
reflect etiological heterogeneity. This estimate can be benchmarked
against the slightly larger weighted mean r
g
between schizophrenia
cohorts of 0.84 (s.e. = 0.05)
21
.
Given the positive evidence of the genetic comparability of these
cohorts, we completed a genome-wide association meta-analysis
of 9.6 million imputed SNPs in 135,458 MDD and major depres
-
sion cases and 344,901 controls (Fig.1). There was no evidence of
residual population stratification
24
(LD score regression intercept
= 1.018, s.e. = 0.009). We estimated
h
SNP
2
to be 8.7% (s.e. = 0.004,
liability scale, assuming lifetime risk of 0.15; Supplementary Fig.1
and Supplementary Table 3b), and note that this is about one-
quarter of the h
2
estimated from twin or family studies
9
. This frac-
tion is somewhat lower than that of other complex traits
18
and is
plausibly due to etiological heterogeneity (reflecting the mean r
g
< 1
between cohorts).
To evaluate the impact of combining major depression cohorts
that used different ascertainment methods, we undertook a series
of genetic risk score (GRS) prediction analyses to demonstrate the
validity of our genome-wide association results for clinical MDD
(Fig. 2). Notably, the variance explained in out-of-sample pre
-
diction increased with the size of the genome-wide association
discovery cohort (Fig.2a), with the GRS from the full discovery
Genome-wide association analyses identify 44
risk variants and refine the genetic architecture of
major depression
A full list of authors and affiliations appears at the end of the paper.
NATURE GENETICS | VOL 50 | MAY 2018 | 668–681 | www.nature.com/naturegenetics
668
© 2018 Nature America Inc., part of Springer Nature. All rights reserved.
Articles
NATure GeNeTics
sample meta-analysis explaining 1.9% of variance in liability
(Fig.2a, Supplementary Fig.2, and Supplementary Table4). For
any randomly selected case and control, GRS ranked cases higher
than controls with probability of 0.57 and the odds ratio of MDD
for those in the tenth versus first GRS decile (OR = 10) was 2.4
(Fig.2b and Supplementary Table4). GRS analyses in other disor
-
ders (for example, schizophrenia
25
) have shown that themean GRS
increases with clinical severity in cases. We found a significantly
higher major depression GRS in those with more severe MDD, as
measured in different ways (Fig.2c). Last, because around one-half
of the major depression cases were identified by self-report (i.e.,
diagnosis or treatment for clinical depression by a medical profess-
ional), we further evaluated the comparability of the 23andMe
cohort with the other cohorts (full meta-analysis excluding
23andMe, ‘FMex23andMe’) as detailed in Fig.2c, Supplementary
Table5, and theSupplementary Note. Taken together, we interpret
these results as supporting this meta-analysis of GWAS for these
seven cohorts.
Implications for the biology of major depression. Our meta-
analysis of seven MDD and major depression cohorts identified 44
independent loci that were statistically significant (P < 5 × 10
−8
), sta-
tistically independent of any other signal
26
, and supported by mul-
tiple SNPs. This number supports our prediction that genome-wide
association discovery in major depression would require about five
times more cases than for schizophrenia (lifetime risk ~1% and h
2
~0.8) to achieve approximately similar power
27
. Of these 44 loci, 30
are new and 14 were significant in a prior study of MDD or depres
-
sive symptoms. The overlap of our findings with prior reports was
as follows: 1 of 1 with CHARGE depressive symptom
14
, 1 of 2 over-
lap with SSGAC depressive symptom
16
, and 12 of 15 overlap with
Hyde et al.
28
. There are few trans-ancestry comparisons for major
depression, so we contrasted these European results with the Han
Chinese CONVERGE study
15
(Supplementary Note). The loci iden-
tified in CONVERGE are uncommon in Europeans (rs12415800,
0.45 versus 0.02; rs35936514, 0.28 versus 0.06) and were not signifi
-
cant in our analysis.
Table2 lists genes in or near the lead SNP in each region, regional
plots are in Supplementary Data1, and Supplementary Tables 6
and 7 provide summaries of available information about the bio-
logical functions of the genes in each region. In theSupplementary
Note, we review four key genes in more detail: OLFM4 and NEGR1
(notable for reported associations with obesity and body mass
index
29–34
), RBFOX1 (notable for independent associations at both
the 5′ and 3′ ends, a splicing regulator
35,36
, with a functional role that
may be consistent with chronic hypothalamic–pituitary–adrenal
axis hyperactivation reported in MDD
37
), and LRFN5 (notable for
its role in presynaptic differentiation
38,39
and neuroinflammation
40
).
Gene-wise analyses identified 153 significant genes after con
-
trolling for multiple comparisons (Supplementary Table7). Many of
these genes were in the extended major histocompatibility complex
(MHC) region (45 of 153), and their interpretation is complicated
by high linkage disequilibrium (LD) and gene density. In addition
to the genes discussed above, other notable and significant genes
outside of the MHC region include multiple potentially ‘druggable’
targets that suggest connections of the pathophysiology of MDD to
neuronal calcium signaling (CACNA1E and CACNA2D1), dopa
-
minergic neurotransmission (DRD2, a principal target of antipsy-
chotics), glutamate neurotransmission (GRIK5 and GRM5), and
presynaptic vesicle trafficking (PCLO).
Finally, comparison of the major depression loci with 108 loci for
schizophrenia
19
identified 6 shared loci. Many SNPs in the extended
MHC region are strongly associated with schizophrenia, but impli
-
cation of the MHC region is new for major depression. Another
example is TCF4 (transcription factor 4), which is strongly asso
-
ciated with schizophrenia but wasnot previouslyassociated with
MDD. TCF4 is essential for normal brain development, and rare
mutations in TCF4 cause Pitt–Hopkins syndrome, which includes
autistic features
41
. The GRS calculated from the schizophrenia
50
45
40
35
30
25
20
15
10
5
0
1234567
Genome-wide significant associations (raw)
389,039
384,951
377,797
343,399
297,997
253,409
228,033
Cumulative major depression cohorts
12
10
8
6
4
2
0
024 6 8 10 12
Expected –log
10
(P)
18
15
12
9
6
3
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
–log
10
(P)
Chromosome
Observed –log
10
(P)
λ = 1.316 n
pvals
= 10,172,741
λ
1,000
= 1.002 (135,458 cases, 344,901 controls)
Fig. 1 | Results of genome-wide association meta-analysis of seven
cohorts for major depression. a, Relationship between adding cohorts
and the number of genome-wide significant genomic regions (before the
vetting used to define the final 44 regions). Beginning with the largest
cohort (cohort 1 on the x axis), we added the next largest cohort (cohort
2) until all cohorts were included (7 cohorts). The number next to each
point is the total effective sample size, equivalent to the sample size where
the numbers of cases and controls are equal. b, Association test quantile–
quantile plot showing a marked departure from the null model of no
associations (y axis truncated at 10
−12
). c, Manhattan plot for association
tests from meta-analysis of 135,458 major depression cases and 344,901
controls, with the x axis showing genomic position (chromosomes 1–22
plus the X chromosome) and the y axis showing statistical significance as
–log
10
(P) z statistics; the threshold for significance accounting for multiple
testing is shown by the red horizontal line (P = 5 × 10
−8
).
NATURE GENETICS | VOL 50 | MAY 2018 | 668–681 | www.nature.com/naturegenetics
669
© 2018 Nature America Inc., part of Springer Nature. All rights reserved.
Articles
NATure GeNeTics
genome-wide association results explained 0.8% of the variance in
liability of MDD (Fig.2c).
Implications from integration of functional genomic data.
Results from ‘omic’ studies of functional features of cells and tis
-
sues are necessary to understand the biological implications of the
results of GWAS for complex disorders
42
. To further elucidate the
biological relevance of the major depression findings, we integrated
the results with functional genomic data. First, using enrichment
analyses, we compared the major depression GWAS findings to
bulk tissue mRNA-seq from Genotype-Tissue Expression (GTEx)
data
43
. Only brain samples showed significant enrichment (Fig.3a),
and the three tissues with the most significant enrichments were all
cortical. Prefrontal cortex and anterior cingulate cortex are impor
-
tant for higher-level executive functions and emotional regulation,
which are often impaired in MDD. Both of these regions were impli
-
cated in a large meta-analysis of brain magnetic resonance imaging
(MRI) findings in adult MDD cases
44
. Second, given the predomi-
nance of neurons in the cortex, we confirmed that the major depres-
sion genetic findings connect to genes expressed in neurons but not
oligodendrocytes or astrocytes (Fig.3b)
45
. Given the different meth-
ods used by the seven MDD/major depression cohorts in this study,
demonstration of enrichment of association signals in the brain
regions expected to be most relevant to MDD provides independent
support for the validity of our approach.
Third, we used partitioned LD score regression
46
to evaluate
the enrichment of the major depression genome-wide associa
-
tion findings in over 50 functional genomic annotations (Fig.3c
and Supplementary Table8). The major finding was the signifi
-
cant enrichment of
h
SNP
2
in genomic regions conserved across 29
Eutherian mammals
47
(20.9-fold enrichment, P = 1.4 × 10
−15
). This
annotation was also the most enriched for schizophrenia
46
. We could
not evaluate regions conserved in primates or human ‘accelerated’
regions, as there were too few for confident evaluation
47
. The other
enrichments implied regulatory activity and included open chro
-
matin in human brain and an epigenetic mark of active enhancers
(H3K4me1). Notably, exonic regions did not show enrichment, sug
-
gesting that, as with schizophrenia
17
, genetic variants that change
exonic sequences may not have a large role in major depression.
We found no evidence that Neanderthal introgressed regions were
enriched for major depression genome-wide association findings
48
.
Fourth, we applied methods to integrate genome-wide associa
-
tion SNP results with those from gene expression and methylation
quantitative trait locus (eQTL and meQTL) studies. SMR
49
analysis
identified 13 major depression–associated SNPs with strong evi
-
dence that they control local gene expression in one or more tissues
and 9 with strong evidence that they control local DNA methylation
(Supplementary Table9 and Supplementary Data2). A transcrip
-
tome-wide association study
50
applied to data from the dorsolateral
prefrontal cortex
51
identified 17 genes where major depression–
associated SNPs influenced gene expression (Supplementary
Table10). These genes included OLFM4 (discussed above).
Fifth, we added additional data types to attempt to improve
understanding of individual loci. For the intergenic associa
-
tions, we evaluated total-stranded RNA-seq data from human
brain and found no evidence for unannotated transcripts in
these regions. A particularly important data type is assessment
of DNA–DNA interactions, which can localize a genome-wide
association finding to a specific gene that may be nearby or hun
-
dreds of kilobases away
52–54
. We integrated the major depression
results with ‘easy Hi-C’ data from brain cortical samples (3 adult,
3 fetal, > 1 billion reads each). These data clarified three asso
-
ciations. The statistically independent associations in NEGR1
(rs1432639, P = 4.6 × 10
−15
) and over 200 kb away (rs12129573,
P = 4.0 × 10
−12
) both implicate NEGR1 (Supplementary Fig.3a),
the former likely due to the presence of a reportedly functional
copy number polymorphism (seeSupplementary Note) and the
presence of intergenic loops. The latter association has evidence
of DNA looping interactions with NEGR1. The association in
SOX5 (rs4074723) and the two statistically independent associa
-
tions in RBFOX1 (rs8063603 and rs7198928, P = 6.9 × 10
−9
and
1.0 × 10
−8
) had only intragenic associations, suggesting that the
genetic variation in the regions of the major depression associa
-
tions act locally and can be assigned to these genes. In contrast,
the association in RERE (rs159963, P = 3.2 × 10
−8
) could not
be assigned to RERE as it may contain super-enhancer elements
given its many DNA–DNA interactions with many nearby genes
(Supplementary Fig.3b).
Implications based on the roles of sets of genes. A parsimonious
explanation for the presence of many significant associations for a
complex trait is that the different associations are part of a higher-
order grouping of genes
55
. These could be a biological pathway or a
collection of genes with a functional connection. Multiple methods
allow evaluation of the connection of major depression genome-
wide association results to sets of genes grouped by empirical or
predicted function (pathway or gene set analysis).
Full pathway analyses are in Supplementary Table11, and 19
pathways with false discovery rate q values < 0.05 are summarized
in Fig.4. The major groupings of significant pathways were as fol
-
lows: RBFOX1, RBFOX2, RBFOX3, or CELF4 regulatory networks;
genes whose mRNAs are bound by FMRP; synaptic genes; genes
Table 1 | Brief description of the seven MDD or major depression cohorts used in the meta-analysis
Sample Country Case ascertainment Cases Controls
PGC29
13,a
Various Structured diagnostic interviews
b
16,823 25,632
deCODE
13
Iceland National inpatient electronic records 1,980 9,536
GenScotland
78,79
UK Structured diagnostic interview 997 6,358
GERA
80
USA Kaiser Permanente Northern California Healthcare electronic medical records
(1995–2013)
7,162 38,307
iPSYCH
81
Denmark National inpatient electronic records 18,629 17,841
UK Biobank
82
(pilot data release) UK From self-reported MDD symptoms or treatment or electronic records
69
14,260 15,480
23andMe
28
(discovery sample)
c
USA Self-reported diagnosis or treatment for clinical depression by a medical
professional
75,607 231,747
Total 135,458 344,901
a
Nineteen samples in addition to the ten samples published in the first PGC-MDD paper
13
.
b
One sample used natural language processing of electronic medical records followed by expert diagnostic review.
c
In Hyde et al.
28
, SNPs in 15 genomic regions met genome-wide significance in the combined discovery and replication samples and 11 regions achieved genome-wide significance in the discovery sample
made available to the research community and used here. More details are provided in Supplementary Tables1–3.
NATURE GENETICS | VOL 50 | MAY 2018 | 668–681 | www.nature.com/naturegenetics
670
© 2018 Nature America Inc., part of Springer Nature. All rights reserved.
Articles
NATure GeNeTics
involved in neuronal morphogenesis; genes involved in neuron pro-
jection; genes associated with schizophrenia (at P < 10
−4
)
19
; genes
involved in central nervous system (CNS) neuron differentiation;
genes encoding voltage-gated calcium channels; genes involved in
cytokine and immune response; and genes known to bind to the ret
-
inoid X receptor. Several of these pathways are implicated by GWAS
of schizophrenia and by rare exonic variation of schizophrenia and
autism
56,57
and immediately suggest shared biological mechanisms
across these disorders.
A key issue for common variant GWAS is their relevance for
pharmacotherapy. We conducted gene set analysis that compared
the major depression genome-wide association results to targets
of antidepressant medications defined by pharmacological stud
-
ies
58
and found that 42 sets of genes encoding proteins bound
by antidepressant medications were highly enriched for smaller
major depression association P values than expected by chance (42
drugs, rank enrichment test P = 8.5 × 10
−10
). This finding connects
our major depression genomic findings to MDD therapeutics and
suggests the salience of these results for new lead compound dis
-
covery for MDD
59
.
Implications based on relationships with other traits. Prior epi
-
demiological studies associated MDD with many other diseases
and traits. Because of limitations inherent to observational stud
-
ies, understanding whether a phenotypic correlation is potentially
causal or if it results from reverse causation or confounding is gen
-
erally difficult. Genetic studies now offer complementary strategies
to assess whether a phenotypic association between MDD and a risk
factor or a comorbidity is mirrored by a nonzero r
g
(common vari-
ant genetic correlation) and, for some of these, evaluate the poten-
tial causality of the association given that exposure to genetic risk
factors begins at conception.
We used LD score regression to estimate the r
g
of major depres-
sion with 221 psychiatric disorders, medical diseases, and human
traits
22,60
. Supplementary Table 12 contains the full results, and
Table3 shows the r
g
values with false discovery rates < 0.01. First,
the r
g
values were very high between our major depression genome-
wide association results and those from two studies of current
depressive symptoms. Both correlations were close to 1 (the samples
in one report overlapped partially with this meta-analysis
16
, but the
samples fromthe other did not
14
).
Second, we found significant positive genetic correlations
between major depression and every psychiatric disorder assessed
along with smoking initiation. This is a comprehensive and well-
powered evaluation of the relationship of MDD with other psychi
-
atric disorders, and these results indicate that the common genetic
0.025
0.020
0.015
0.010
0.005
0.000
Variance explained on liability scale
Decile odds ratio
Target sample
Ta rget sample
PGC29-LOO
iPSYCH
PGC29-LOO
iPSYCH
NESDA
1.9 × 10
–8
6.8 × 10
–102
5.1 × 10
–77
4.0 × 10
–2
5.0 × 10
–34
1.8 × 10
–3
9.1 × 10
–47
1.1 × 10
–2
4.1 × 10
–4
1.3 × 10
–38
2.3 × 10
–26
8.6 × 10
–23
3.2 × 10
–72
1.5 × 10
–132
2.8 × 10
–137
Discovery GWAS
Decile
PGC29-LOO
FMex23andMe
23andMe
Full meta
scz
Early AAO
Late AAO
Moderate
Severe
Single
Recurrent
Stage II
Stage IV
Stage III
2.5
2.0
1.5
1.0
12345678910
Secondary phenotype
1.6
1.5
1.4
1.3
1.2
1.1
OR per s.d. of MDD GRS
P = 7.9 × 10
–6
P = 0.045
P = 0.022
P = 9.8 × 10
–4
ab
c
Target sample
Münster
PGC29-LOO
iPSYCH
Fig. 2 | Genetic risk score prediction analyses into PGC29 MDD target samples. a, Variance explained (liability scale) based on different discovery
samples for three target samples: PGC29 (16,823 cases, 25,632 controls), iPSYCH (a nationally representative sample of 18,629 cases and 17,841
controls), and a clinical cohort from Münster not included in the genome-wide association analysis (845 MDD inpatient cases, 834 controls). For PGC29-
LOO, the target sample was one of the PGC29 samples with the discovery sample being the remaining 28 PGC29 samples, hence representing leave-
one-out(LOO) analysis. b, Odds ratios of major depression per GRS decile relative to the first decile for the iPSYCH and PGC29 target samples. c, Odds
ratios of major depression in GRS s.d.: PGC29-LOO,3,950 early-onset versus 3,950 late-onset cases; PGC29-LOO,4,958 severe versus 3,976 moderate
cases defined by count of endorsed MDD symptom criteria; iPSYCH,5,574 cases of recurrent MDD versus 12,968 single-episode cases; andNESDA from
PGC29, severity defined as chronic/unremitting MDD, 610 ‘stage IV’ cases versus 499 ‘stage II’ or 332 first-episode MDD cases
77
. Error bars represent
95% confidence intervals. Logistic regression association test P values are shown in the target sample for the GRS generated from SNPs with P < 0.05 in
the discovery sample.FMex23andMe, full meta-analysis excluding 23andMe; scz, schizophrenia
19
.
NATURE GENETICS | VOL 50 | MAY 2018 | 668–681 | www.nature.com/naturegenetics
671
© 2018 Nature America Inc., part of Springer Nature. All rights reserved.
