CoChat

The DOI matches its Crossref record; 40 of the 40 references it cites were found in a registry or an index; 12 of the 33 sources its claims cite were checked for support.

Using polygenic scores to assess liability to antisocial behavior

Open access
Kazantseva, A. V., Yakovleva, D. V., Davydova, Yu. D., Khusnutdinova, E. K. · 2025 · Vavilov Journal of Genetics and Breeding

Abstract

PGS alone explained a small proportion of variance in liability to ASB (1.1-1.5%), while the inclusion of social parameters increased variance explained.

Overview

CoChat Summary

TL;DR

This study tested whether polygenic scores (PGS) derived from European antisocial-behavior GWAS data predict severe criminal behavior (homicide) in a Russian cohort, and whether adding social factors improves prediction. PGS alone explained little variance, while models including social factors explained substantially more, with the best model reaching 21.2% variance explained in offenders with a family history of mental illness.

Key findings

  • In 227 homicide offenders and 254 healthy donors from the Volga-Ural region, PGS based on four nominally associated SNPs explained a small proportion of variance in liability to antisocial behavior: 0.9% in the total group, 0.9% in proactive aggression, 0.9% with comorbid substance abuse, and 1.5% with family history of mental illness.
  • Adding social predictors increased variance explained to 14.5% (total ASB), 15.8% (proactive ASB), 21.0% (ASB with comorbid addiction), and 21.2% (ASB with family history of mental illness); the best model had r² = 21% and AUC = 0.802 for subjects with family history of mental disorders.
  • Significant social predictors included traumatic brain injury or severe chronic disease in anamnesis (β = 1.4, p = 9.8 × 10⁻⁹) and present smoking (β = 1.7, p = 1.4 × 10⁻⁷), while past smoking was associated with decreased ASB liability (β = –1.3, p = 8.8 × 10⁻⁴).
  • Weighted and unweighted PGS using all ten SNPs explained 1.1–1.5% and 1.0–1.3% of variance, respectively; with social predictors, variance explained was 16.2–21.2% for weighted PGS and 16.1–21.0% for unweighted PGS.

Methods

  • Case-control replication study in the Volga-Ural region of Russia: 227 criminal offenders convicted of homicide and 254 healthy donors matched by age, ethnicity, and gender; offenders were mainly men (93%) with mean age 41.5 ± 14.5 years, and the sample was 48% Russians, 34.8% Tatars, and 17.2% Bashkirs.
  • Ten “top” SNPs from a prior ASB GWAS meta-analysis (CADM2, REV3L, FOXP1, FOXP2, BDNF, FURIN, XKR6, TMEM18, SORCS3, ZIC4) were genotyped by real-time PCR with KASP chemistry; PGS were calculated as weighted and unweighted sums using effect estimates from Karlsson Linnér et al. (2021).
  • Multiple logistic regression models included PGS alone and then PGS plus social factors selected by backward selection; models were compared by lowest p-value, highest Nagelkerke pseudo-R², and highest AUC.

Limitations

  • The authors state that the SNP set used for PGS was small, which may explain the low variance attributed to genetic effects; they also note that only biallelic polymorphisms were studied, leaving tandem repeats and microdeletions/duplications unexamined.
  • The authors state the sample size is small, which may produce type I and II errors and requires future enlargement; they also note that only a limited number of social factors were analyzed, excluding child-parent relationship, criminal organization membership, physical or sexual violence, social isolation, and personality type.
  • The authors state that most offenders had excessive alcohol/opiate use, so they cannot rule out whether findings are attributable to heavy alcohol drinking; the text also makes evident that the PGS models rely on European GWAS effect estimates and no Russian ASB GWAS was available.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.18699/vjgb-25-91

Cites

40 of 40 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Table 1. Examined top SNPs linked to antisocial behavior: data from previous ASB GWAS and the VUR cohort
Table 1. Examined top SNPs linked to antisocial behavior: data from previous ASB GWAS and the VUR cohortCC BY · PubMed Central
Table 2. Characteristics of the examined groups of criminal offenders and healthy donors and analysis of differences in social factors between the groups Note. Statistically significant differences between the groups based on p-value < 0.05 are shown in bold. TBI/disease – traumatic brain injury or severe chronic disease in anamnesis. Dashes indicate non-applicable data.
Table 2. Characteristics of the examined groups of criminal offenders and healthy donors and analysis of differences in social factors between the groups Note. Statistically significant differences between the groups based on p-value < 0.05 are shown in bold. TBI/disease – traumatic brain injury or severe chronic disease in anamnesis. Dashes indicate non-applicable data.CC BY · PubMed Central
Table 3. Regression models of liability to ASB based on a polygenic score and social factors as predictors Note. ASB – antisocial behavior; FamPsy ASB – ASB in individuals with a family history of mental illness; β – regression coefficient for each predictor in the model; SE – standard error of β; AUC – area under curve; TBI/disease – traumatic brain injury or severe chronic disease in anamnesis. PGS_0.1 was based on effect estimates for REV3L rs458806, FOXP1 rs11720703, XKR6 rs4240671, and SORCS3 rs11596214; PGS_W and PGS_UW were PGS based on “weighted” and “unweighted” effect estimates for ten SNPs, correspondingly.
Table 3. Regression models of liability to ASB based on a polygenic score and social factors as predictors Note. ASB – antisocial behavior; FamPsy ASB – ASB in individuals with a family history of mental illness; β – regression coefficient for each predictor in the model; SE – standard error of β; AUC – area under curve; TBI/disease – traumatic brain injury or severe chronic disease in anamnesis. PGS_0.1 was based on effect estimates for REV3L rs458806, FOXP1 rs11720703, XKR6 rs4240671, and SORCS3 rs11596214; PGS_W and PGS_UW were PGS based on “weighted” and “unweighted” effect estimates for ten SNPs, correspondingly.CC BY · PubMed Central
Fig. 1. Proportion of variance (adjusted r2) in liability to antisocial behavior in the examined cohort explained by predictors included in multiple regression models based on PGS calculation of SNPs with p < 0.1 (PGS_0.1), “weighted effects” of all SNPs (PGS_all_W), “unweighted effects” of all SNPs (PGS_all_UW) with inclusion of social predictors (PGS_0.1 + cov, PGS_all_W + cov, PGS_all_UW + cov). Examined groups of ASB: total group of homicide offenders (ASB); homicide offenders with a proactive type of ASB (Proactive ASB), family history of mental illness or criminal behavior (FamPsy ASB), or substance abuse (Addictive ASB).
Fig. 1. Proportion of variance (adjusted r2) in liability to antisocial behavior in the examined cohort explained by predictors included in multiple regression models based on PGS calculation of SNPs with p < 0.1 (PGS_0.1), “weighted effects” of all SNPs (PGS_all_W), “unweighted effects” of all SNPs (PGS_all_UW) with inclusion of social predictors (PGS_0.1 + cov, PGS_all_W + cov, PGS_all_UW + cov). Examined groups of ASB: total group of homicide offenders (ASB); homicide offenders with a proactive type of ASB (Proactive ASB), family history of mental illness or criminal behavior (FamPsy ASB), or substance abuse (Addictive ASB).CC BY · PubMed Central
Fig. 2. ROC curves and areas under the curves (AUCs) for various regression models predicting liability to manifest ASB in individuals with a family history of mental illness or criminal behavior (abbreviations are given as in Fig. 1).
Fig. 2. ROC curves and areas under the curves (AUCs) for various regression models predicting liability to manifest ASB in individuals with a family history of mental illness or criminal behavior (abbreviations are given as in Fig. 1).CC BY · PubMed Central

Verification

Claims the paper makes

8 Valid · 1 Not valid · 13 Not checked · 39 Own results and interpretations

32 more claims

Create a free account to see all 61 claims and to verify the 4 references still open.

See all claims — free account
  • Probably not a substantial claim. 14
  • It may state two separate claims as one. 15

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

1 of 1

Citations

Cites

Show all 40 references

Research works printed but not linked: 3

  1. Ip H.F., Van der Laan C.M., Krapohl E.M.L., Brikell I., SánchezMora C., Nolte I.M., St Pourcain B., … Lundström S., Plomin R., Bartels M., Nivard M.G., Boomsma D.I. Genetic association study of childhood aggression across raters, instruments, and age. Transl Psychiatry. 2021;11(1):413. doi 10.1038/s41398-021-01480-x · DOI · Paper page
  2. Karlsson Linnér R., Mallard T.T., Barr P.B., Sanchez-Roige S., Mado­ le J.W., Driver M.N., Poore H.E., … Waldman I.D., Palmer A.A., Harden K.P., Koellinger P.D., Dick D.M. Multivariate analysis of 1.5 million people identifies genetic associations with traits related to self-regulation and addiction. Nat Neurosci. 2021;24(10):1367- 1376. doi 10.1038/s41593-021-00908-3 · DOI · Paper page
  3. Odintsova V.V., Hagenbeek F.A., van der Laan C.M., van de Weijer S., Boomsma D.I. Genetics and epigenetics of human aggression. Handb Clin Neurol. 2023;197:13-44. doi 10.1016/B978-0-12-821375-9.00005-0 · DOI

Metadata

Authors
Kazantseva, A. V., Yakovleva, D. V., Davydova, Y. D., & Khusnutdinova, E. K.
Affiliations
  • Russian Academy of Sciences (RU)
  • Institute of Biochemistry and Genetics – Subdivision of the Ufa Federal Research Centre of the Russian Academy of Sciences
Venue
Vavilov Journal of Genetics and Breeding
Published
2025-10-09
Volume
29
Issue
6
Pages
838-846
ISSN
  • 1814-554X (ISSN-L)
  • 2500-3259
OpenAlex
W4415734481
Cited by
0
References
40
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Vavilov Journal of Genetics and Breeding (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
Ministry of Science and Higher Education of the Russian Federation
Topics
  • Social Sciences › Psychology › Clinical Psychology › Psychopathy, Forensic Psychiatry, Sexual Offending
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Genetics › Genetic Associations and Epidemiology
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Neurotransmitter Receptor Influence on Behavior
Keywords
polygenic risk score, antisocial behavior, homicide offenders, genetic risk factors, single nucleotide polymorphisms, social factors, traumatic brain injury, substance use comorbidity, family history of mental illness, Russian population, Volga-Ural region, CADM2, brain-derived neurotrophic factor, FOXP1, FOXP2, ZIC4, SORCS3, TMEM18, furin, replication study

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Using polygenic scores to assess liability to antisocial behavior. https://app.cochat.ai/papers/10.18699/vjgb-25-91
The paper
Kazantseva, A. V., Yakovleva, D. V., Davydova, Y. D., & Khusnutdinova, E. K. (2025). Using polygenic scores to assess liability to antisocial behavior. Vavilov Journal of Genetics and Breeding. https://doi.org/10.18699/vjgb-25-91

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.