Archival Report
Examining Sex-Differentiated Genetic Effects
Across Neuropsychiatric and Behavioral Traits
Joanna Martin, Ekaterina A. Khramtsova, Slavina B. Goleva, Gabriëlla A.M. Blokland,
Michela Traglia, Raymond K. Walters, Christopher Hübel, Jonathan R.I. Coleman,
Gerome Breen, Anders D. Børglum, Ditte Demontis, Jakob Grove, Thomas Werge,
Janita Bralten, Cynthia M. Bulik, Phil H. Lee, Carol A. Mathews, Roseann E. Peterson,
Stacey J. Winham, Naomi Wray, Howard J. Edenberg, Wei Guo, Yin Yao, Benjamin M. Neale,
Stephen V. Faraone, Tracey L. Petryshen, Lauren A. Weiss, Laramie E. Duncan,
Jill M. Goldstein, Jordan W. Smoller, Barbara E. Stranger, and Lea K. Davis, on behalf of the Sex
Differences Cross-Disorder Analysis Group of the Psychiatric Genomics Consortium
ABSTRACT
BACKGROUND: The origin of sex differences in prevalence and presentation of neuropsychiatric and behavioral traits
is largely unknown. Given established genetic contributions and correlations, we tested for a sex-differentiated
genetic architecture within and between traits.
METHODS: Using European ancestry genome-wide association summary statistics for 20 neuropsychiatric and
behavioral traits, we tested for sex differences in single nucleotide polymorphism (SNP)-based heritability and
genetic correlation (r
g
, 1). For each trait, we computed per-SNP z scores from sex-stratified regression
coefficients and identified genes with sex-differentiated effects using a gene-based approach. We calculated
correlation coefficients between z scores to test for shared sex-differentiated effects. Finally, we tested for sex
differences in across-trait genetic correlations.
RESULTS: We observed no consistent sex differences in SNP-based heritability. Between-sex, within-trait genetic
correlations were high, although ,1 for educational attainment and risk-taking behavior. We identified 4 genes
with significant sex-differentiated effects across 3 traits. Several trait pairs shared sex-differentiated effects. The
top genes with sex-differentiated effects were enriched for multiple gene sets, including neuron- and synapse-
related sets. Most between-trait genetic correlation estimates were not significantly different between sexes, with
exceptions (educational attainment and risk-taking behavior).
CONCLUSIONS: Sex differences in the common autosomal genetic architecture of neuropsychiatric and behavioral
phenotypes are small and polygenic and unlikely to fully account for observed sex-differentiated attributes. Larger
sample sizes are needed to identify sex-differentiated effects for most traits. For well-powered studies, we
identified genes with sex-differentiated effects that were enriched for neuron-related and other biological
functions. This work motivates further investigation of genetic and environmental influences on sex differences.
https://doi.org/10.1016/j.biopsych.2020.12.024
Despite widespread evidence of sex differences across human
complex traits, including neuropsychiatric and behavioral
phenotypes (1), the etiology of these differences remains
poorly understood. Accumulating evidence suggests that sex
differences in complex human phenotypes are likely to include
an autosomal genetic component beyond that contributed by
sex chromosomes (2–5). Understanding the biological basis of
sex differences in human disease, including neuropsychiatric
phenotypes, is critical for developing sex-informed diagnostics
and therapeutics and realizing the promise of precision medi-
cine (4). Moreover, genetic variants with sex-differentiated ef-
fects across multiple traits may influence patterns of
comorbidity for neuropsychiatric and related behavioral
phenotypes, suggesting the need for cross-disorder genetic
analyses to be evaluated in the context of sex-differentiated
effects (6–11).
Neuropsychiatric and behavioral phenotypes are generally
characterized by a complex and highly polygenic etiology (12).
Many of these traits share common genetic risk variants
(13,14). Specific genetic loci with pleiotropic effects are known
to impact risk for multiple related phenotypes (12). However, it
is not yet known whether these pleiotropic effects are
consistent across sex.
Several recent studies have investigated sex-differentiated
genetic effects for a number of neuropsychiatric traits
(15–29). Given evidence of phenotypic sex differences in
SEE COMMENTARY ON PAGE e63
ª 2021 Society of Biological Psychiatry. This is an open access article under the
CC BY license (http://creativecommons.org/licenses/by/4.0/).
1127
ISSN: 0006-3223 Biological Psychiatry June 15, 2021; 89:1127 –1137 www.sobp.org/journal
Biological
Psychiatry
prevalence and presentation as well as genetic correlations
between these traits (13), we aimed to systematically test the
hypothesis that neuropsychiatric and behavioral phenotypes
have a partially sex-differentiated autosomal genetic archi-
tecture that may be shared across traits. In this study, we have
characterized the 1) sex-dependent genetic architecture for a
range of neuropsychiatric and behavioral traits, 2) degree of
shared genetic architecture between males and females within
each phenotype, and 3) sex-specific patterns of genetic effects
shared across traits.
METHODS AND MATERIALS
Datasets
We collected sex-stratified genome-wide association study
(GWAS) meta-analysis summary statistics for 20 neuropsy-
chiatric and behavioral traits (Table 1; see Sex-Stratified
Datasets in Supplement 1), chosen based on data availabil-
ity. See Table S1 in Supplement 2 for information about data
availability. We used a broad definition of brain-based human
complex traits, given the overwhelming evidence of shared
genetic effects across such traits (13). We used results from
European ancestry GWASs only to minimize any bias that may
arise from ancestry differences and because large sex-
stratified GWAS summary statistics from other ancestries are
not currently available. We analyzed autosomal-only common
variants with a minor allele frequency .1%.
Sex-Specific Single Nucleotide Polymorphism–
Based Heritability
For each trait, we calculated sex-specific observed scale
single nucleotide polymorphism (SNP)-based heritability
(SNP-h
2
) using linkage disequilibrium (LD) score regression
(LDSC) with precomputed European ancestry LD scores
Table 1. Summary of Analyzed Datasets of Neuropsychiatric and Behavioral Traits
Phenotype (Full Name) Acronym
Female
Cases (n)
Female
Controls (n)
Male
Cases (n)
Male
Controls (n)
M:F Case
Ratio Sample Type
Consortium/
Group Reference
Attention-Deficit/
Hyperactivity Disorder
ADHD 4945 16,246 14,154 17,948 2.86 Clinical case-control PGC1iPSYCH (15)
Alcohol Dependence ALCD 2504 6033 5932 9412 2.37 Clinical case-control PGC (16)
Anxiety Disorders ANX 3148 191,005 1813 165,175 0.58 General population (UK) Neale
laboratory
(17)
Autism Spectrum
Disorder
ASD 7498 24,309 30,168 32,417 4.02 Clinical case-control PGC1iPSYCH (18,19)
Bipolar Disorder BD 10,753 14,225 7331 13,572 0.68 Clinical case-control PGC2 (20)
Cannabis Use (Ever) CUE 17,244 71,742 17,414 50,737 1.01 General population (UK) N/A N/A
Insomnia INS 19,521 39,846 12,863 40,776 0.66 General population (UK) N/A (21)
Major Depressive
Disorder
MDD 10,711 11,745 5021 11,226 0.47 Clinical and population
case-control
PGC1 (20)
Major Depressive
Disorder
N/A
a
13,492 180,661 7156 159,832 0.53 General population (UK) Neale
laboratory
(17)
Major Depressive
Disorder Recurrent
MDDR 6026 8949 2643 8162 0.44 Clinical case-control PGC1 (20)
Obsessive-Compulsive
Disorder
OCD 1525 4307 1249 2789 0.82 Clinical case-control PGC (22)
Posttraumatic Stress
Disorder
PTSD 968 2457 585 4025 0.60 Clinical case-control PGC (23)
Risk-Taking Behavior RTB 32,285 143,678 51,392 100,984 1.59 General population (UK) N/A (24)
Schizophrenia SCZ 9837 16,763 18,346 17,122 1.86 Clinica l case-control PGC2 (20)
Smoking (Current) SMKC 16,995 176,392 20,093 146,226 1.18 General population (UK) Neale
laboratory
(17)
Smoking (Previous) SMKP 62,305 131,082 65,245 101,074 1.05 General population (UK) Neale
laboratory
(17)
Females (n) Males (n)
Alcohol Use ALCC 59,088 53,088 0.90 General population (UK) (25)
Alcohol Use N/A
a
85,800 55,120 0.64 General population (26)
Age at First Birth AFB 189,656 48,408 0.26 General population (27)
Educational Attainment EA 182,286 146,631 0.80 General population (28)
Number of Children
Ever Born
NEB 225,230 103,909 0.46 General population (27)
Neuroticism NEU 144,660 142,875 0.99 General population (UK) (29)
F, female; M, male; N/A, not applicable; PGC, Psychiatric Genomics Consortium; UK, United Kingdom.
a
These summary statistics were not used for analysis (see Sex-Stratified Datasets in Supplement 1 for details).
Sex-Differentiated Genetic Effects on Complex Traits
1128 Biological Psychiatry June 15, 2021; 89:1127–1137 www.sobp.org/journal
Biological
Psychiatry
(excluding SNPs in the HLA/MHC [human leukocyte antigen/
major histocompatibility complex] region; chr6:25-34M) (30).
For 11 binary traits, we also estimated liability scale SNP-h
2
,
using sex-specific population prevalence rates from two
sources, as described below. For comparison with this pri-
mary analysis, we also used a second method, LDAK-
SumHer (31), to estimate SNP-h
2
, using the LD-adjusted
kinships (LDAK) heritability model.
We obtained sex-specific trait prevalence estimates from
the United States (32) and cumulative incidence rates from
Denmark (33) to compare the SNP-h
2
estimates using two
different sources of information. See Sex-Specific Trait
Prevalences for Estimating SNP-h
2
in Supplement 1 and
Tables S2 and S3 in Supplement 2 for details.
For traits with nonzero SNP- h
2
estimates (i.e., where con-
fidence intervals did not overlap with zero) in both sexes, we
tested whether sex-specific SNP-h
2
estimates were signifi-
cantly different by calculating z scores using equation 1 (below)
and obtaining corresponding p values from a normal distribu-
tion. We corrected for multiple tests using Bonferroni (n =12
independent tests for n = 5 continuous traits and n = 7 binary
traits with nonzero liability scale SNP-h
2
in both sexes; p =
.0042).
z 2 score ¼
STAT
female
2STAT
male
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
SE
2
female
1SE
2
male
q
(1)
In equation 1, STAT can be any statistic for which we want
to assess the difference between the sexes, including SNP-h
2
,
r
g
, and GWAS
b
values; SE is the standard error for the sta-
tistic. This test is well calibrated when STAT/SE is normally
distributed and the test statistics are independent between
sexes and is conservative if the statistics are positively
correlated.
Genetic Correlation
We used LDSC to estimate genetic correlations (r
g
) 1) be-
tween sexes, within each trait, and 2) between each trait pair,
within sex (Figure 1A). For between-sex, within-trait correla-
tions, we tested the null hypothesis that r
g
, 1 using a
1-tailed test compared with a normal distribution (z =(12 r
g
)/
SE). We applied a Bonferroni multiple-testing correction (p ,
.0031 based on 16 traits). Next, we tested whether the
between-trait r
g
estimates were different for males (r
gM
) and
females (r
gF
) by using a z score approximation based on
block jackknife to estimate the standard error of r
gM
2 r
gF
in
LDSC. As with other LDSC analyses, this approach is robust
to sample overlap. We applied a false discovery rate multiple-
testing correction.
Between-Sex, Within-Trait Genetic Heterogeneity
Given that only summary statistics from sex-stratified GWASs
were available, the analysis of sex-differentiated genetic
Figure 1. (A) Schematic illustration of the key analyses used to investigate between-sex, within-trait and between-trait, within-sex differences. (B–D)
Estimates of sex stratified SNP-based heritability (SNP-h
2
)on(B) the observed scale for continuous traits and the liability scale using population prevalence
based on (C) Denmark (DK) and (D) the United States (US). Estimates were obtained from linkage disequilibrium score regression. Points represent the
estimated SNP-h
2
in males (blue) and females (red), while bars represent SE of the SNP-h
2
estimates. Significant sex difference in heritability is denoted as
follows: *p , .0042 (adjusted p value threshold corrected for multiple testing using Bonferroni). #Traits for which significance in difference is not interpretable
owing to negative or nonsignificant from zero SNP-h
2
value for one of the measurements. (E) Within-trait, between-sex genetic correlation (r
g
) estimates using
linkage disequilibrium score regression. Points represent the estimated r
g
, and bars represent SE of the r
g
estimates. Significant deviation from 1 is denoted as
follows: *p , .0031 (adjusted p value threshold corrected for multiple testing using Bonferroni). ADHD, attention-deficit/hyperactivity disorder; AFB, age at first
birth; ALCC, alcohol use; ALCD, alcohol dependence; ANX, anxiety disorders; ASD, autism spectrum disorder; BD, bipolar disorder; CUE, cannabis use (ever);
EA, educational attainment; INS, insomnia; MDD, major depressive disorder; MDDR, major depressive disorder recurrent; NEB, number of children ever born;
NEU, neuroticism; OCD, obsessive-compulsive disorder; PTSD, posttraumatic stress disorder; RTB, risk-taking behavior; SCZ, schizophrenia; SMKC, smoking
(current); SMKP, smoking (previous); SNP, single nucleotide polymorphism.
Sex-Differentiated Genetic Effects on Complex Traits
Biological Psyc hiatry June 15, 2021; 89:1127–1137 www.sobp.org/journal 1129
Biological
Psychiatry
effects was limited to the following z score approach. For each
SNP in the sex-stratified GWAS of each trait, we assessed
between-sex, within-trait heterogeneity using z scores (which
are correlated with Cochran’s Q statistic but provide direc-
tionality of the effect) as in equation 1. This test quantifies the
sex difference in SNP association effect size, similar to,
although not the same as, an interaction test (34).
Sharing of Variants With Sex-Differentiated Effects
Across Traits
To assess which traits share sex-differentiated effects (i.e.,
variants at the extreme ends of the z score distribution), we
assessed the Pearson correlation coefficient between z scores
(i.e., the differences of
b
values from male-only and female-
only GWASs) for pairs of traits. Given that there are many
nonindependent observations, owing to SNPs in LD, we used a
block jackknife approach to estimate the significance of the
Pearson correlation (35,36). SNPs were assigned to 1 of 1000
contiguous blocks based on genomic position. For each trait
pair, Pearson ’s correlation was calculated on the full set of z
scores and then recalculated after each block was removed,
thus estimating the jackknife error and p values.
Gene-Based Analysis, Differential Gene
Expression, and Gene-Set Enrichment Analysis of
Genes With Sex-Differentiated Effects
We used the Functional Mapping and Annotation of GWAS
(FUMA) SNP2GENE web tool (37) to perform gene-based
analysis using MAGMA v1.08 (38,39). We examined whether
the genes exhibiting a genome-wide significant sex difference
(from MAGMA) demonstrate sex-differentiated gene expres-
sion in brain tissues from the Genotype-Tissue Expression
project v8 (https://www.gtexportal.org/home/datasets)(20).
After mapping SNPs to genes (using a default window size of
0), we performed gene set enrichment analysis on the union
(across phenotypes) of genes with sex-differentiated effects
using GSEA (https://software.broadinstitute.org/gsea/index.
jsp). See Gene-Based Analysis, Differential Gene Expression,
and Gene-Set Enrichment Analysis of Genes With Sex-
Differentiated Effects in Supplement 1 for details.
RESULTS
Sex-Stratified SNP-h
2
Estimates
Sex-specific SNP-h
2
estimates using LDSC are presented in
Figure 1B–D, with details provided in Table S4 in Supplement 2.
Several traits (posttraumatic stress disorder and recurrent
major depressive disorder [MDD] in males and autism spec-
trum disorder (ASD) and alcohol dependence in females) did
not have sufficient power (or had excessive heterogeneity) and
we did not detect a polygenic signal, and therefore sex dif-
ferences could not be assessed. Thus, although we report sex
difference estimates for all traits in Table S4 in Supplement 2,
these cannot be reliably interpreted for these 4 traits, as one of
the sexes exhibited a near-zero SNP-h
2
estimate. The liability
scale SNP-h
2
estimates using population prevalence from the
United States and cumulative incidence from Denmark were
highly correlated (r = .97, p = 4.7 3 10
210
)(Figure S1 in
Supplement 1). Age at first birth was the only trait with a
significant (after multiple testing correction; p , .0042) sex
difference in SNP-h
2
estimates (females: SNP-h
2
= 0.052, SE =
0.004; males: SNP-h
2
= 0.113, SE = 0.010; z score = 25.81,
p = 6.43 3 10
29
).
Observed scale SNP-h
2
estimates based on LDAK-SumHer
were somewhat higher than the estimates obtained in LDSC
and moderately correlated with them (r = .69, p = 8.5 3 10
27
for all traits; r = .85, p = 3.3 3 10
211
excluding the 4 traits for
which SNP-h
2
could not be reliably estimated in LDSC); see
Table S5 in Supplement 2 and Figures S1 and S2 in
Supplement 1 for details. Higher estimates from the LDAK
model relative to the LDSC model have been previously
observed (31,38). In contrast to LDSC results, age at first birth
did not show a significant sex difference (z score = 1.94, p =
.052), with an effect in the opposite direction to that observed
using LDSC. Using LDAK, the liability scale (adjusted based on
each population) SNP-h
2
estimates differed by sex for the
following traits: recurrent MDD (United States: z score = 24.68,
p = 2.84 3 10
26
; Denmark: z score = 24.46, p = 8.06 3 10
26
),
ASD (United States: z score = 2.94, p = .0033; Denmark: z
score = 3.28, p = .0011), and schizophrenia (Denmark: z
score = 23.16, p = .0016). These results were not observed
using LDSC, and indeed SNP-h
2
could not be estimated reli-
ably in LDSC for ASD in females or recurrent MDD in males.
The biggest discrepancies between estimates obtained from
LDSC and LDAK were for the traits with the smallest sample
sizes ( Figure S3 in Supplement 1). The SNP-h
2
results for
attention-deficit/hyperactivity disorder (ADHD) and ASD were
similar, albeit somewhat higher, for both LDSC and LDAK
when using estimates based on a Danish child-specific study
(39) compared with using prevalence estimates from the whole
Danish population (Tables S4 and S5 in Supplement 2)(33).
Between-Sex, Within-Trait Genetic Correlation
Analysis
We quantified the genetic correlation between males and fe-
males for each trait (excluding the 4 traits where SNP-h
2
could
not be estimated in one of the sexes) (Figure 1E and Table S6
in Supplement 2). We found moderate-to-high genetic corre-
lations for all traits (r
g
= 0.68–1.21); these all differed signifi-
cantly from zero, and we also detected a significant difference
from 1 for risk-taking behavior (r
g
= 0.81, SE = 0.04) and
educational attainment (r
g
= 0.92, SE = 0.02), after correcting
for multiple tests (p , .0031), suggesting a modest degree of
common variant heterogeneity in males and females for these
phenotypes.
Between-Sex, Within-Trait Heterogeneity Across
Variants
To assess sex differences in genetic effects of individual
common variants, for each trait we computed z scores and
corresponding p values for each SNP, using equation 1.
Figure S4 in Supplement 1 shows the quantile-quantile plots of
the z score p values for all traits. While there were no genome-
wide significant (p , 5 3 10
28
) differences between male and
female
b
values for any individual SNP, we observed deviation
from the expected null distribution (Figure S4 in Supplement 1)
for ADHD, lifetime cannabis use, MDD, number of children
born, and schizophrenia. Figure 2A shows a Miami plot for
Sex-Differentiated Genetic Effects on Complex Traits
1130 Biological Psychiatry June 15, 2021; 89:1127–1137 www.sobp.org/journal
Biological
Psychiatry
