Leading Edge
Review
Defining the Genetic, Genomic, Cellular, and
Diagnostic Architectures of Psychiatric Disorders
Patrick F. Sullivan1,2,* and Daniel H. Geschwind3,4,*
1Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
2Departments of Genetics and Psychiatry, University of North Carolina, Chapel Hill, NC, USA
3Departments of Neurology, Psychiatry, and Human Genetics, David Geffen School of Medicine, University of California, Los Angeles,
CA, USA
4Institute for Precision Health, David Geffen School of Medicine, University of California, Los Angeles, CA, USA
*Correspondence: (P.F.S.), (D.H.G.)
https://doi.org/10.1016/j.cell.2019.01.015
Studies of the genetics of psychiatric disorders have become one of the most exciting and fast-
moving areas in human genetics. A decade ago, there were few reproducible findings, and now
there are hundreds. In this review, we focus on the findings that have illuminated the genetic archi-
tecture of psychiatric disorders and the challenges of using these findings to inform our
understanding of pathophysiology. The evidence is now overwhelming that psychiatric disorders
are ‘‘polygenic’’—that many genetic loci contribute to risk. With the exception of a subset
of those with ASD, few individuals with a psychiatric disorder have a single, deterministic
genetic cause; rather, developing a psychiatric disorder is influenced by hundreds of different ge-
netic variants, consistent with a polygenic model. As progressively larger studies have uncovered
more about their genetic architecture, the need to elucidate additional architectures has become
clear. Even if we were to have complete knowledge of the genetic architecture of a psychiatric dis-
order, full understanding requires deep knowledge of the functional genomic architecture—the
implicated loci impact regulatory processes that influence gene expression and the functional co-
ordination of genes that control biological processes. Following from this is cellular architecture: of
all brain regions, cell types, and developmental stages, where and when are the functional architec-
tures operative? Given that the genetic architectures of different psychiatric disorders often
strongly overlap, we are challenged to re-evaluate and refine the diagnostic architectures of psychi-
atric disorders using fundamental genetic and neurobiological data.
Introduction nings of psychiatric disorders with numerous findings that
Psychiatric disorders are the most enigmatic maladies in medi- meet modern criteria for significance and reproducibility
cine. Although their existence has been known for millennia (Geschwind and Flint, 2015; Sullivan et al., 2018). In this
(Porter, 2002) and their impact on public health well-docu- Review, we focus on the findings that have illuminated the
mented, remarkably little is known about their causal risk factors ‘‘genetic architecture’’ of psychiatric disorders and the chal-
and fundamental neurobiology despite a considerable corpus of lenges of using these findings to inform our understanding of
research. In the past century, many have applied the best tools pathophysiology. Genetic architecture refers to the overall
then available but, until recently, without reproducible suc- composition of the implicated risk variants in the population—
cesses. The lack of success using approaches that were fruitful the total number of variants and, for each, the frequencies in
elsewhere is attributable an inadequate toolkit and the intrinsic those afflicted and in the general population and the degree
complexity of the brain. Psychiatric disorders impact higher of risk conferred (Timpson et al., 2018). The concept of genetic
cortical functions (mood, behavior, perception, and cognition), architecture is applicable to any trait (e.g., Huntington’s disease
which are far more difficult to localize, quantify, and model is caused by a rare, deterministic variant). Knowledge of
than more basic neurological functions. In addition, psychiatric genetic architecture can help optimize gene discovery (e.g.,
disorders are defined based on self-report and observation of study design, ascertainment, choice of genotyping technology)
cognition and behavior rather than on direct measurement of (Timpson et al., 2018; Visscher et al., 2012). Genetic architec-
an etiological factor, making them syndromes rather than single ture can inform prospects for clinical utility: although many
diseases. These features strongly suggest diverse and complex deterministic monogenetic conditions are predicted or diag-
etiologies. nosed using genetic testing, application to most psychiatric
Despite these challenges, there has been remarkable prog- disorders traverses far more murky, probabilistic terrain (Timp-
ress in the past decade in elucidating the genetic underpin- son et al., 2018).
162 Cell 177, March 21, 2019 ª 2019 Elsevier Inc.
,Figure 1. Relationship of the Levels of Disease Architecture to Different Stages of Analysis
Genetic studies identify the loci and causal variants that impact disease and thereby its genetic architecture. The subset of causal variants in coding regions are
typically directly assignable to genes. As many loci are non-coding, regulatory regions and the genes they regulate need to be empirically defined and identified—
such studies render the functional architecture of disease. As psychiatric disorders all appear to be polygenic, it is also necessary to consider the implicated
genes in the context of biological networks and pathways. Sets of genes and networks can be placed in specific developmental stages and cell types to generate
more precise understanding of their effects on brain regions and circuits. Diagnostic architecture—the structure of the interrelationships between psychiatric
syndromes—is subsequently refined by increased knowledge at each of these levels.
The evidence is now overwhelming that psychiatric disorders and what circuits do they influence? Finally, the data used to
have a ‘‘polygenic’’ basis—that many genetic loci, mostly with diagnose psychiatric disorders consist of signs and symptoms
small effect sizes, contribute to risk (Visscher et al., 2017). In determined during patient-clinician interactions that infre-
this respect, psychiatric disorders are broadly similar to other quently have recourse to objective biomarkers to support or
common biomedical diseases. The polygenic concept allows refute a diagnosis. Furthermore, the internationally accepted
for the fact that some individuals can harbor genetic variants of definitions of psychiatric disorders were crafted by experts
far larger effects. This is particularly salient for autism spectrum and influenced by traditions dating back a century or more.
disorder (ASD), where a large effect variant is present in 15% Given that the genetic architectures of different psychiatric
of cases, along with smaller proportions of individuals with disorders can strongly overlap, we are challenged to re-eval-
Tourette’s syndrome (TS), attention-deficit hyperactivity disor- uate and refine the diagnostic architectures of psychiatric
der (ADHD), and schizophrenia (SCZ) (Iossifov et al., 2012; disorders with respect to fundamental genetic and neurobio-
Sanders et al., 2012; Satterstrom et al., 2018b; Singh et al., logical data.
2016; Willsey et al., 2017). A polygenic model can include
weak and strong genetic effects, as well as non-genetic influ- Psychiatric Disorders and Genetics
ences (e.g., the impact of environmental exposures and life Definitions
events [e.g., chronic fear], the impact of individual choices). A Many psychiatric disorders are internationally recognized (World
key empirical finding is that genetic risk can be non-specific Health Organization, 1993). In this Review, we focus on the 10
and shared to varying extents across many adult- and child- psychiatric disorders that have been the subject of the greatest
hood-onset psychiatric disorders (Antilla et al., 2018; Cross-Dis- scrutiny by geneticists and all are the focus of working groups in
order Group of the Psychiatric Genomics Consortium, 2013; the Psychiatric Genomics Consortium (PGC) (Sullivan et al.,
Schork et al., 2019). 2018). We do not cover dementia and intellectual disability (ID),
As progressively larger studies of psychiatric disorders have which are often considered neurological conditions with promi-
uncovered increasingly more about their genetic architecture, nent psychiatric manifestations, but recognize the inherent
the need to elucidate additional architectures has become arbitrariness of following this conventional delineation. Table 1
clear (Figure 1). Even if we were to have complete knowledge contains brief definitions of each condition, along with lifetime
of the genetic architecture of a psychiatric disorder, full under- prevalence rates and twin heritabilities. The essence of each dis-
standing requires deep knowledge of the functional genomic order is a persistent, pervasive, and pathological pattern of
architecture—how these loci interact in the nucleus (often abnormal mood (as in mania or major depression), perception
across large distances), how gene and isoform expression (e.g., auditory hallucinations in SCZ, bizarrely distorted body
are coordinated for many genes, and how these affect net- image in anorexia nervosa [AN]), behavior (e.g., repetitive
works. Second, following from this, is cellular architecture— hand-washing in obsessive-compulsive disorder [OCD], inju-
of all brain regions, cell types, and developmental stages, rious ethanol consumption in alcohol dependence [ALC]), or
where and when are the functional architectures operative, higher-level cognition (e.g., delusions in SCZ). People with
Cell 177, March 21, 2019 163
, 164 Cell 177, March 21, 2019
Table 1. Descriptive Features of 10 Psychiatric Disorders
Lifetim Twin SNP GWA GWAS
Abbreviation Name Prevalence Heritability Heritability Cases Loci Essential Characteristics Notable Impacts
ADHD attention-deficit 0.053 0.76 0.216 20,183 12 persistent inattention, costs estimated at
hyperactivity disorder hyperactivity, impulsivity $100 billion USD per year
ALC alcohol dependence 0.125 0.51 0.090 14,904 2 persistent ethanol use most expensive psychiatric
despite tolerance, disorder (total costs > $225
withdrawal, dysfunction billion/year)
AN anorexia nervosa 0.009 0.58 0.110 16,992 8 dangerously low weight notably high standardized
from self-starvation mortality ratio
ASD autism spectrum 0.017 0.74 0.118 18,381 5 abnormal social interaction wide range of function, from
disorder and communication complete care to exceptional
beginning before age 3 achievement
BIP bipolar disorder 0.010 0.85 0.213 29,764 30 manic-depressive illness, nearly as disabling as SCZ
episodes of mania usually
with depressive episodes
MDD major depressive 0.162 0.37 0.087 135,458 44 unipolar depression, top five in burden of disease
disorder persistent dysphoria with globally
physical/cognitive
symptoms
OCD obsessive-compulsive 0.011 0.47 0.280 2,688 0 uncontrollable, persistent top 10 globally for lost income
disorder thoughts (obsessions) and and decreased quality of life
repetitive behaviors
(compulsions)
PTSD post-traumatic stress 0.068 0.46 0.038 23,212 2 trauma-related high medical and psychiatric
disorder re-experiencing, avoidance, comorbidities (suicide,
negative thoughts, and substance depdence)
hyperarousala
SCZ schizophrenia 0.004 0.81 0.244 40,675 145 longstanding delusions life expectancy decreased
and hallucinations by 12–15 years
TS Tourette’s syndrome 0.005 0.37b 0.350 4,819 1 vocal or motor tics (stereotyped, comorbid psychiatric
involuntary movement disorders cause more disability
and utterances) than tics
All definitions are made more restrictive by requiring persistence over time (e.g., the criteria for SCZ require R6 months of symptoms), presence in different contexts (e.g., for ADHD, inattention at
home, school, and in peer interactions), and significant impairment. See Table S1 for data and citations. Updated from Sullivan et al., (2012).
a
PTSD is distinctive in requiring traumatic exposure to death, injury, or sexual violence.
b
Heritability from national pedigrees is higher (0.77).
Review
Defining the Genetic, Genomic, Cellular, and
Diagnostic Architectures of Psychiatric Disorders
Patrick F. Sullivan1,2,* and Daniel H. Geschwind3,4,*
1Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, Stockholm, Sweden
2Departments of Genetics and Psychiatry, University of North Carolina, Chapel Hill, NC, USA
3Departments of Neurology, Psychiatry, and Human Genetics, David Geffen School of Medicine, University of California, Los Angeles,
CA, USA
4Institute for Precision Health, David Geffen School of Medicine, University of California, Los Angeles, CA, USA
*Correspondence: (P.F.S.), (D.H.G.)
https://doi.org/10.1016/j.cell.2019.01.015
Studies of the genetics of psychiatric disorders have become one of the most exciting and fast-
moving areas in human genetics. A decade ago, there were few reproducible findings, and now
there are hundreds. In this review, we focus on the findings that have illuminated the genetic archi-
tecture of psychiatric disorders and the challenges of using these findings to inform our
understanding of pathophysiology. The evidence is now overwhelming that psychiatric disorders
are ‘‘polygenic’’—that many genetic loci contribute to risk. With the exception of a subset
of those with ASD, few individuals with a psychiatric disorder have a single, deterministic
genetic cause; rather, developing a psychiatric disorder is influenced by hundreds of different ge-
netic variants, consistent with a polygenic model. As progressively larger studies have uncovered
more about their genetic architecture, the need to elucidate additional architectures has become
clear. Even if we were to have complete knowledge of the genetic architecture of a psychiatric dis-
order, full understanding requires deep knowledge of the functional genomic architecture—the
implicated loci impact regulatory processes that influence gene expression and the functional co-
ordination of genes that control biological processes. Following from this is cellular architecture: of
all brain regions, cell types, and developmental stages, where and when are the functional architec-
tures operative? Given that the genetic architectures of different psychiatric disorders often
strongly overlap, we are challenged to re-evaluate and refine the diagnostic architectures of psychi-
atric disorders using fundamental genetic and neurobiological data.
Introduction nings of psychiatric disorders with numerous findings that
Psychiatric disorders are the most enigmatic maladies in medi- meet modern criteria for significance and reproducibility
cine. Although their existence has been known for millennia (Geschwind and Flint, 2015; Sullivan et al., 2018). In this
(Porter, 2002) and their impact on public health well-docu- Review, we focus on the findings that have illuminated the
mented, remarkably little is known about their causal risk factors ‘‘genetic architecture’’ of psychiatric disorders and the chal-
and fundamental neurobiology despite a considerable corpus of lenges of using these findings to inform our understanding of
research. In the past century, many have applied the best tools pathophysiology. Genetic architecture refers to the overall
then available but, until recently, without reproducible suc- composition of the implicated risk variants in the population—
cesses. The lack of success using approaches that were fruitful the total number of variants and, for each, the frequencies in
elsewhere is attributable an inadequate toolkit and the intrinsic those afflicted and in the general population and the degree
complexity of the brain. Psychiatric disorders impact higher of risk conferred (Timpson et al., 2018). The concept of genetic
cortical functions (mood, behavior, perception, and cognition), architecture is applicable to any trait (e.g., Huntington’s disease
which are far more difficult to localize, quantify, and model is caused by a rare, deterministic variant). Knowledge of
than more basic neurological functions. In addition, psychiatric genetic architecture can help optimize gene discovery (e.g.,
disorders are defined based on self-report and observation of study design, ascertainment, choice of genotyping technology)
cognition and behavior rather than on direct measurement of (Timpson et al., 2018; Visscher et al., 2012). Genetic architec-
an etiological factor, making them syndromes rather than single ture can inform prospects for clinical utility: although many
diseases. These features strongly suggest diverse and complex deterministic monogenetic conditions are predicted or diag-
etiologies. nosed using genetic testing, application to most psychiatric
Despite these challenges, there has been remarkable prog- disorders traverses far more murky, probabilistic terrain (Timp-
ress in the past decade in elucidating the genetic underpin- son et al., 2018).
162 Cell 177, March 21, 2019 ª 2019 Elsevier Inc.
,Figure 1. Relationship of the Levels of Disease Architecture to Different Stages of Analysis
Genetic studies identify the loci and causal variants that impact disease and thereby its genetic architecture. The subset of causal variants in coding regions are
typically directly assignable to genes. As many loci are non-coding, regulatory regions and the genes they regulate need to be empirically defined and identified—
such studies render the functional architecture of disease. As psychiatric disorders all appear to be polygenic, it is also necessary to consider the implicated
genes in the context of biological networks and pathways. Sets of genes and networks can be placed in specific developmental stages and cell types to generate
more precise understanding of their effects on brain regions and circuits. Diagnostic architecture—the structure of the interrelationships between psychiatric
syndromes—is subsequently refined by increased knowledge at each of these levels.
The evidence is now overwhelming that psychiatric disorders and what circuits do they influence? Finally, the data used to
have a ‘‘polygenic’’ basis—that many genetic loci, mostly with diagnose psychiatric disorders consist of signs and symptoms
small effect sizes, contribute to risk (Visscher et al., 2017). In determined during patient-clinician interactions that infre-
this respect, psychiatric disorders are broadly similar to other quently have recourse to objective biomarkers to support or
common biomedical diseases. The polygenic concept allows refute a diagnosis. Furthermore, the internationally accepted
for the fact that some individuals can harbor genetic variants of definitions of psychiatric disorders were crafted by experts
far larger effects. This is particularly salient for autism spectrum and influenced by traditions dating back a century or more.
disorder (ASD), where a large effect variant is present in 15% Given that the genetic architectures of different psychiatric
of cases, along with smaller proportions of individuals with disorders can strongly overlap, we are challenged to re-eval-
Tourette’s syndrome (TS), attention-deficit hyperactivity disor- uate and refine the diagnostic architectures of psychiatric
der (ADHD), and schizophrenia (SCZ) (Iossifov et al., 2012; disorders with respect to fundamental genetic and neurobio-
Sanders et al., 2012; Satterstrom et al., 2018b; Singh et al., logical data.
2016; Willsey et al., 2017). A polygenic model can include
weak and strong genetic effects, as well as non-genetic influ- Psychiatric Disorders and Genetics
ences (e.g., the impact of environmental exposures and life Definitions
events [e.g., chronic fear], the impact of individual choices). A Many psychiatric disorders are internationally recognized (World
key empirical finding is that genetic risk can be non-specific Health Organization, 1993). In this Review, we focus on the 10
and shared to varying extents across many adult- and child- psychiatric disorders that have been the subject of the greatest
hood-onset psychiatric disorders (Antilla et al., 2018; Cross-Dis- scrutiny by geneticists and all are the focus of working groups in
order Group of the Psychiatric Genomics Consortium, 2013; the Psychiatric Genomics Consortium (PGC) (Sullivan et al.,
Schork et al., 2019). 2018). We do not cover dementia and intellectual disability (ID),
As progressively larger studies of psychiatric disorders have which are often considered neurological conditions with promi-
uncovered increasingly more about their genetic architecture, nent psychiatric manifestations, but recognize the inherent
the need to elucidate additional architectures has become arbitrariness of following this conventional delineation. Table 1
clear (Figure 1). Even if we were to have complete knowledge contains brief definitions of each condition, along with lifetime
of the genetic architecture of a psychiatric disorder, full under- prevalence rates and twin heritabilities. The essence of each dis-
standing requires deep knowledge of the functional genomic order is a persistent, pervasive, and pathological pattern of
architecture—how these loci interact in the nucleus (often abnormal mood (as in mania or major depression), perception
across large distances), how gene and isoform expression (e.g., auditory hallucinations in SCZ, bizarrely distorted body
are coordinated for many genes, and how these affect net- image in anorexia nervosa [AN]), behavior (e.g., repetitive
works. Second, following from this, is cellular architecture— hand-washing in obsessive-compulsive disorder [OCD], inju-
of all brain regions, cell types, and developmental stages, rious ethanol consumption in alcohol dependence [ALC]), or
where and when are the functional architectures operative, higher-level cognition (e.g., delusions in SCZ). People with
Cell 177, March 21, 2019 163
, 164 Cell 177, March 21, 2019
Table 1. Descriptive Features of 10 Psychiatric Disorders
Lifetim Twin SNP GWA GWAS
Abbreviation Name Prevalence Heritability Heritability Cases Loci Essential Characteristics Notable Impacts
ADHD attention-deficit 0.053 0.76 0.216 20,183 12 persistent inattention, costs estimated at
hyperactivity disorder hyperactivity, impulsivity $100 billion USD per year
ALC alcohol dependence 0.125 0.51 0.090 14,904 2 persistent ethanol use most expensive psychiatric
despite tolerance, disorder (total costs > $225
withdrawal, dysfunction billion/year)
AN anorexia nervosa 0.009 0.58 0.110 16,992 8 dangerously low weight notably high standardized
from self-starvation mortality ratio
ASD autism spectrum 0.017 0.74 0.118 18,381 5 abnormal social interaction wide range of function, from
disorder and communication complete care to exceptional
beginning before age 3 achievement
BIP bipolar disorder 0.010 0.85 0.213 29,764 30 manic-depressive illness, nearly as disabling as SCZ
episodes of mania usually
with depressive episodes
MDD major depressive 0.162 0.37 0.087 135,458 44 unipolar depression, top five in burden of disease
disorder persistent dysphoria with globally
physical/cognitive
symptoms
OCD obsessive-compulsive 0.011 0.47 0.280 2,688 0 uncontrollable, persistent top 10 globally for lost income
disorder thoughts (obsessions) and and decreased quality of life
repetitive behaviors
(compulsions)
PTSD post-traumatic stress 0.068 0.46 0.038 23,212 2 trauma-related high medical and psychiatric
disorder re-experiencing, avoidance, comorbidities (suicide,
negative thoughts, and substance depdence)
hyperarousala
SCZ schizophrenia 0.004 0.81 0.244 40,675 145 longstanding delusions life expectancy decreased
and hallucinations by 12–15 years
TS Tourette’s syndrome 0.005 0.37b 0.350 4,819 1 vocal or motor tics (stereotyped, comorbid psychiatric
involuntary movement disorders cause more disability
and utterances) than tics
All definitions are made more restrictive by requiring persistence over time (e.g., the criteria for SCZ require R6 months of symptoms), presence in different contexts (e.g., for ADHD, inattention at
home, school, and in peer interactions), and significant impairment. See Table S1 for data and citations. Updated from Sullivan et al., (2012).
a
PTSD is distinctive in requiring traumatic exposure to death, injury, or sexual violence.
b
Heritability from national pedigrees is higher (0.77).
