Skip to main content

The first Externalizing Consortium GWAS publication (EXT 1.0; Karlsson Linnér et al. (2021). Multivariate analysis of 1.5 million people identifies genetic associations with traits related to self-regulation and addiction. Nat. Neurosci. 24, 1367–1376.) included summary statistics contributed by 23andMe. Accordingly, before we can provide summary statistics from that original externalizing GWAS, you must obtain permission from 23andMe.  To obtain the EXT 1.0 GWAS summary statistics, please follow the procedures described below carefully.


Procedures to obtain EXT 1.0 summary stats (including 23andMe), as reported in Karlsson Linnér et al. (2021):

STEP 1 : (a) Fill out this form to request access to the externalizing GWAS summary statistics from the principal investigators of the Externalizing Consortium, AND (b) initiate a request with 23andMe through the 23andMe Publication Dataset Access Procedure. In your 23andMe data access request, please follow the below instructions (failure to do so will cause delays):

  1. From the list of Behavior GWAS, select ‘Cannabis Use (Pasman, 2018)’ and ‘Tobacco and Alcohol Use (Liu, 2019)’.
  2. When asked “Were you able to select all of the GWAS you’re interested in receiving”, reply ‘No’.
  3. When asked, “Is there anything else you would like to tell us about the GWAS you are requesting”, copy and paste the following statement: ‘We request permission to work with the externalizing GWAS summary statistics from Karlsson Linnér et al. Multivariate analysis of 1.5 million people identifies genetic associations with traits related to self-regulation and addiction. Nat. Neurosci. 24, 1367–1376 (2021), which utilized GWAS summary statistics from the following two studies that contained contributing data from 23andMe: Liu et al., Association studies of up to 1.2 million individuals yield new insights into the genetic etiology of tobacco and alcohol use. Nat. Genet. 51, 237–244 (2019); Pasman et al., GWAS of lifetime cannabis use reveals new risk loci, genetic overlap with psychiatric traits, and a causal influence of schizophrenia. Nat. Neurosci. 21, 1161–1170 (2018).’


Note that you do not need to obtain any actual summary statistics from 23andMe, you are only requesting permission to obtain the externalizing GWAS summary statistics from us.

STEP 2 : Once your institution has signed a Data Transfer Agreement and Statement of Work, you will receive an email from giving you permission to receive the externalizing summary stats. Forward this request to Emily Balcke at and Richard Karlsson Linnér at We will then provide the externalizing GWAS summary statistics.

STEP 3 : All publications using the externalizing summary statistics should be sent to 23andMe, with copy to, to confirm that the topic of the paper is in the scope of the project as described in the Statement of Work with 23andMe and that the 23andMe research participant acknowledgment and Externalizing Consortium acknowledgment has been included.


Procedures to obtain EXT 1.0 summary stats (excluding 23andMe):

To simplify the sharing process, we reran the analyses from Karlsson Linnér et al. (2021), removing the 23andMe samples. To obtain the EXT-23andme summary statistics, we applied the EXT1.0 genomic structural equation model to the summary statistics of GWASs on externalizing behaviors and disorders that exclude 23andMe data. The EXT-23andMe model structure, fit, and parameters were similar to those of the original analysis; and the EXT1.0 and EXT-23andMe factors had similar genetic correlations with external traits. The variance in externalizing behaviors explained by the polygenic score for externalizing decreased from 8.9% to  8.4% for COGA and from 10.5% to 8.6% for Add Health when using the EXT-23andMe summary statistics to calculate polygenic scores. Those analyses are reported in Williams et al. (2023). To obtain these summary statistics (EXT–23andMe), please complete this form.

Note that all papers using these summary statistics must cite the original Externalizing GWAS manuscript and the manuscript describing the reduced model:

  • Karlsson Linnér et al. Multivariate analysis of 1.5 million people identifies genetic associations with traits related to self-regulation and addiction. Nat. Neurosci. 24, 1367–1376 (2021).
  • Williams et al. Guidelines for Evaluating the Comparability of Down-Sampled GWAS Summary Statistics. Behav. Genet. 53, 404-415 (2023).


Existing Data Access Requests


PI: Alemany, S. 
Vall d’Hebron Research Institute (VHIR)
This project aims to examine whether genetic susceptibility to externalizing behavior moderates associations between environmental risk factors and the phenotypic manifestation of psychopathology.

PI: Andreasson, O.
Institution: University of Oslo
Project: The externalizing GWAS summary statistics will be used in ongoing projects directed to apply new statistical methods and uncover more granular genetic architecture of psychiatric traits. Moreover, the connection with comorbid traits and disorders will be investigated with hopes to identify specific molecular mechanisms involved in their pathogenesis.

PI: Bacanu, S.- A.
Institution: Virginia Commonwealth University
Project: The study will investigate transcriptomic and proteomic signals, and will use in MTAG to improve the power to detect signals for psychiatric conditions that have an externalizing component, e.g. AUD and PTSD.

PI: Barker, E. D.
Institution: King’s College London
Project: Creation of a polygenic risk score to examine if the score can help explain common phenotypic comorbidities for early onset and persistent conduct problem children in the Avon Longitudinal Study of Parents and Children.

PI: Belangero, S.I.N.
Institution: Universidade Federal de São Paulo
Project: This project aims to investigate the genetic association and genetic causation linking traumatic experience to psychopathology (internalizing and externalizing factors).

PI: Biroli, P.
Institution: University of Zurich
Project:  Estimation of a genetically-informed structural model of the process of cognitive and socio-emotional development of children in the first five years of life using a large longitudinal survey of children born in the UK at the turn of the millennium. The externalizing GWAS summary statistics will be used to build a corresponding polygenic risk variable for the survey participants and subsequently employed in the aforementioned structural model.

PI: Boivin, M.
Institution: Université Laval
Project: This project will use the summary statistics to compute a polygenic index for externalizing behaviors (extPGI) using PRS-CS for the participants from the Quebec Newborn Twin Study and Quebec Longitudinal Study of Child Development. The combined contributions of extPGI and measured environments will be tested to participants’ developmental outcomes, including aggression, ADHD, anxiety, depression, substance use, educational achievement, cortisol, and health outcomes.

PI: Bralten, J.
Institution: Radboud University Medical Center
Project:The research project will investigate the genetic and neurobiological mechanisms underlying self-regulation and development in adolescents as part of the Growing Up Together in Society (GUTS) consortium

PI: Breen, G.
Institution: King’s College London
Project: Polygenic scoring in the Genetic Links to Anxiety and Depression Study and the PGC Eating Disorders phase 3.

PI: Brennan, C.
Institution: Queen Mary University of London
Project: An exploration of causality between impulsivity and mania in UK Biobank.

PI: Cantave, Y. C.
University of Groningen
This study aims to examine whether early experiences of adversity interact with genetic factors, including the externalizing PGS, to predict youth’s report of supportive and caring relationships with their parents and friends.

PI: Coombes, B.
Institution: Mayo Clinic
Project:The summary statistics for externalizing will be used to create PGS for a range of different studies in psychiatry at the Mayo Clinic. Specifically, they are interested in the association of this PGS with substance use disorders.

PI: Derks, E.
Institution: QIMR Berghofer
Project: This project aims to delineate the causal influences between cannabis use and schizophrenia and explore the potential mediating role of externalizing behavior.

PI: Dinkelbach, L.
Institution: University Hospital Essen, University Duisberg-Essen
Project: This project aims to analyse the relationship between hormones as well as puberty timing and the occurrence of mental health issues.

PI: Docherty, A.
Institution: University of Utah
Project: Examination of SNP-based genetic correlations and conditional GWAS results using the Externalizing PRS and GWAS data on death by suicide.

PI: Elliott, A.
Institution: Massachusetts General Hospital
Project: This project aims to use the externalizing GWAS summary statistics as a comparison phenotype in the investigation of genetic liability contributing to different depression phenotypes.

PI: Fullerton, J.
Institution: Neuroscience Research Australia
Project: Calculation of polygenic risk scores in a cohort of patients with diagnoses of bipolar disorder (n=1575) to determine if there are unique genetic signatures which can distinguish these clinical subgroups within the broader bipolar disorder phenotype.

PI: Ganna, A.
Institution: University of Helsinki
Project: Estimation of the effects of polygenic risk scores, including the Externalizing polygenic risk score, on the risk of 72 major diseases, and estimation of attributable disability-adjusted life-years lost/gained due to belonging in different percentiles of the polygenic scores.

PI: Hartman, C.
Institution: University Medical Center Groningen
Project: This study aims to investigate if an unhealthy diet exacerbates the genetic risk of impulsivity that is a symptom of ADHD. In addition, the study will explore if an unhealthy diet interacts with other lifestyle behaviors (i.e. physical activity, sleep, smoking, drinking) to exacerbate the genetic risk of impulsivity. The analyses will be conducted in the Dutch Lifelines cohort which is a prospective population-based cohort study of 167,729 persons from the north of the Netherlands.

PI: Holla, B.
Institution: National Institute of Mental Health and Neurosciences, Bengaluru
Project: The main aim of this study is to examine the predictive accuracy of PRS derived from European GWAS summary statistics on SAS/Indian samples, as well as to identify factors associated with differential prediction accuracy.

PI: Howe, L. 
University of Bristol
The main aim of this study is to assess the role of genetic confounding in the association between screen time and mental health, and in the association between childhood adversity and mental health.

PI: Hulshoff Pol, H.
Institution: Utrecht University
Project: This project will use the externalizing GWAS summary statistics as discovery data in polygenic risk score analyses in an ongoing study on externalizing-related behaviors and brain development in the Adolescent Brain Cognitive Development (ABCD) study.

PI: Kaprio, J.
Institution: University of Helsinki
Project: Creation of polygenic risk scores excluding all Finns and PheWas analysis in FinnGen.

PI: Kember, R.
Institution: University of Pennsylvania
Project: The present study involves sex-stratified analyses of externalizing polygenic scores across multiple sites from the psycheMERGE network.

PI: Konzok, J.
Institution: University of Regensburg
Project: The aim of this project is to replicate and validate the results from a Mendelian Randomization analysis on childhood maltreatment and externalizing disorders, including a common externalizing factor restricted on ICD coded phenotypes.

PI: Kozhemaiko, N.
Institution: Brigham and Women’s Hospital, Harvard Medical School
Project: The present study aims to test associations between externalizing PGS and sleep metrics derived from sleep EEG in a large target cohort of healthy participants.

PI: Kretschmer, T.
Institution: University of Groningen
Project: The present study combines questionnaire data with genome-wide SNP variations to investigate the extent to which phenotypic associations between bullying involvement and later internalizing and externalizing problems may be explained by individual differences in genetic vulnerability to internalizing and externalizing problems.

PI: Kretschmer, T.
Institution: University of Groningen
Project: The study investigates gene-environment interplay in the development of externalizing problems from childhood through adulthood. Evocative gene-environment correlations and genetic nurture are tested in TRacking Adolescents’ Individual Lives Survey.

PI: Lee, J. J.
Institution: University of Minnesota
Project: This project aims to build EXT PGS for individuals in the Minnesota twin and adoption samples, and use these scores to predict life outcomes.

PI: Lee, P. H.
Institution: Broad Institute
Project: This project aims to leverage multivariate methods to better define the shared and disorder-specific basis of brain disorders and co-occurring traits including externalizing problems.

PI: Li, J.
Institution: University of Wisconsin-Madison
Project: This study proposes to examine associations between externalizing polygenic scores and trajectories of externalizing phenotypes in large, publicly available datasets (e.g., Add Health, ABCD). This study is a collaboration of members of the HiTOP Genomics Working Group (chaired by Monika Waszczuk and Katherine Jonas).

PI: Liu, B.
Institution: Beijing Normal University
Project: This project will use the externalizing GWAS summary statistics to calculate polygenic risk scores in other data sets. In addition, in combination with brain imaging data, the project will explore the relationship between these polygenic risk scores and brain structure and function.

PI: Liu, D.
Institution: Chinese Academy of Sciences
Project: This study aims to examine whether externalizing traits are causally associated with stroke and its subtypes and Alzheimer’s disease and dementia.

PI: Maher, B.
Institution: Johns Hopkins University
Project: This project will examine whether polygenic risk for externalizing behaviors is associated with drug use behaviors and suicide attempt in diverse ancestral populations.

PI: Martin, H.
Institution: Wellcome Sanger Institute
Project: This project will use two UK birth cohorts with SNP array and whole exome sequence data to investigate the temporally dynamic genetic architectures of cognitive performance and internalising and externalising behaviour.

PI: McAdams, T.
Institution: King’s College London
Project: This project aims to explore the direction of effects between parenting behaviours and offspring depressive symptoms: an intergenerational mendelian randomisation-direction of causation study (MR-DOC).

PI: Mehta, D.
Institution: Queensland University of Technology
Project: This project aims to address the impact of maternal behavior and environmental stress on children’s health outcomes including mental health and non-cognitive skills through genetic and epigenetic pathways. The team will examine the impact of the externalizing behaviors through the prediction of a polygenic risk score for externalizing behaviors for ALSPAC mothers using the externalizing GWAS summary statistics and genetic data of ALSPAC mothers. They will then examine the association of this risk score with various child health outcomes including depression, behavior disorders and developmental problems.

PI: Orri, M.
McGill University
The study team will calculate a polygenic score to estimate the association between this score and several phenotypes of aggression measured during adolescence in a population-based sample.

PI: Pasaniuc, B.
Institution: University of California – Los Angeles
Project: The methods from the externalizing GWAS paper will be applied to the UCLA ATLAS biobank data and connected with EHR-derived features.

PI: Pasman, J.
Institution: Karolinska Institute
Project: This project aims to perform an investigation of the genetic relationship between major depressive disorder and a broad range of risk factors and outcomes, including externalizing behavior.

PI: Pfenning, A.
Institution: Carnegie Mellon University
Project: Expansion of existing analyses in addiction-associated traits to fine-map variants to GWAS loci and infer cell type and brain region specificity with a functional epigenomics approach using cutting-edge machine learning techniques.

PI: Pingault, J-B.
Institution: University College London
Project: We plan to investigate intergenerational associations between polygenic scores, including the externalizing PGS, and child conduct problems to dissect genetic transmission from potential genetic nurture effects.

PI: Rovaris, D. L.
Institution: Universidade de São Paulo
Project: We are investigating the biological connection between cortisol dysregulation and ADHD. After conducting initial analyses, we aim to validate a potential mechanism elucidating the link between cortisol variation and ADHD through externalizing comorbidities. Consequently, we intend to employ Mendelian Randomization models to assess whether there is a causal link between cortisol and externalizing behaviors beyond ADHD.

PI: Runze, J.
Institution: University of Amsterdam
Project: We plan to investigate the role of polygenic scores and the parenting environment in child disruptive behavior

PI: Savage, J.
Institution: Vrije Universiteit Amsterdam
Project: This project will investigate the overlap between externalizing and heterogeneous dimension/subgroups of alcohol misuse. The externalizing summary statistics will be used for genetic correlation and Mendelian randomization analyses.

PI: Shen, X.
University of Edinburgh
This project aims to create polygenic risk scores of externalising behaviour in large-scale cohorts, such as Generation Scotland, ABCD and ALSPAC to identify brain and environmental factors that may associate with genetic proxy of externalising behaviours.

PI: Seidlitz, J.
Institution: University of Pennsylvania
Project: This research project seeks to investigate the genetic and transcriptomic overlap of neuropsychiatric and neurodegenerative disorders through GWAS, TWAS, and postmortem RNA sequencing to better understand the relationships between disease nosology and dimensions of disease pathology.

PI: Silveira, P. P.
Institution: McGill University
Project: This research project seeks to identify functional genomic markers of vulnerability to develop substance use disorders (SUD). The externalizing GWAS summary statistics will be used to select overlapping genes between a selected co-expressed network associated with impulsivity and the SNPs annotated to genes from the GWAS. A Spearman’s correlation analysis will be performed to evaluate the correlation between the module membership of each gene within our network and the p-value for the significance of the SNPs from the GWAS study.

PI: Stallings, M. C.
Institution: University of Colorado – Boulder
Project: The study team plans to parse externalizing and sensation-seeking behavior into three latent factors: substance use/abuse, risky sexual behaviors, and non-sexual/non-substance behaviors related to externalizing and self-regulation. They plan to explore the variance in sexual, substance use related, and mental health measures the PGS they construct for each of the three factors explained in a replication sample– the Center for Antisocial Drug Dependence (CADD), and whether the percentage of variance explained changes from adolescence to adulthood.

PI: Stein, M.
Institution: University of California – San Diego
Project: Creation of a PRS to be tested in STARRS and TRACK-TBI cohorts to see if they predict various mental health and functional outcomes.

PI: Straub, V.
Institution: University of Oxford
Project: This project aims to apply the externalizing GWAS summary statistics to the Our Future Health Cohort data in a phenome-wide analysis of the correlates of risk for disinhibited health risk behaviors.

PI: Streit, F.
Institution: Central Institute of Mental Health – Mannheim
Project: Assessment of genetic correlations of Borderline Personality Disorder (BPD) with a broad range of risk factors, and other psychiatric disorders and phenotypes. The study team will also calculate parallel analyses with other related mental disorders (e.g. ADHD, PTSD (PGC)), to compare the correlational patterns with the one from BPD.

PI: Su, J.
Institution: Arizona State University
Project: This research project aims to examine how genetic predispositions and social environmental factors (e.g., family, peer, and cultural environment) interact to influence alcohol use and related outcomes among adolescents in diverse populations. The study team plans to use the externalizing GWAS summary statistics to calculate polygenic scores (PRS) in the Adolescent Brain Cognitive Development (ABCD) study sample and examine the role of risk and protective environmental factors in moderating and/or mediating genetic risk among racially and ethnically diverse youth.

PI: Thongsima, S.
Institution:National Science and Technology Development Agency (National Biobank of Thailand)
Project: We plan to do Genetic Predisposition and Network Analysis of Cannabis Use Disorder (CUD) Severity in the Thai Population. The GWAS summary statistics will be used to construct and validate PRS for Thai people.

PI: Toikumo, S.
University of Pennsylvania
This research project aims to use the externalizing GWAS summary statistics to explore the potential causal relationship between Tobacco Use Disorder (TUD) and externalizing traits. The PI previously performed a genome-wide association study of TUD across five major biobanks using the TUD phecode (318) and identified tobacco-related loci with some novel loci.

PI: Tropf, F.
Institution: Crest-ENSAE
Project: Using data from the 1958 birth cohort study on (non-)cognitive skills, socioeconomic life outcomes and neighborhood deprivation, alongside genetic data, this project will investigate the pathways of (non-)cognitive polygenic scores, including the externalizing PRS, through measured behaviors and skills to socioeconomic life outcomes as well as to test for gene-environment interactions.

PI: Tubbs, J.
Institution: Massachusetts General Hospital
Project: The research team plans to estimate the genetic correlation between externalizing, latent cognitive factors, and latent mood disorder factors.

PI: Tucker-Drob, E.
Institution: University of Texas at Austin
Project: The study team are examining genetic correlations between externalizing and the Big Five personality domains as part of a new GWAS of the Big Five.

PI: Uher, R.
Institution: Dalhousie University
Project: This project aims to characterize the clinical and genetic predictors of major mood disorders, psychotic disorders, and suicidal behaviour in youth at low and high familial risk.

PI: Voloudakis, G.
Institution: Icahn School of Medicine at Mount Sinai
Project: This project will assess the genetic similarity between neuropsychiatric phenotypes, with an emphasis on binge eating disorder and externalizing behaviors.

PI: Wang, F.
Institution: University of Pittsburgh
Project: The study team will create an externalizing PRS and examine how that influences externalizing behaviors over time in adolescents from families oversampled for alcohol use disorder.

PI: White, T.
Institution: National Institute of Mental Health
Project: The study team will investigate the relationship between the externalizing polygenic score and symptom concordance in monozygotic (MZ) twin pairs.

PI: Ystrom, E.
Institution: University of Oslo
Project: The study team is making a database of polygenic scores for a range of different psychiatric and behavioral traits, to use in a variety of different projects.