Table 1. Summary of genetic studies in obsessive-compulsive disorder (OCD) cited throughout this review.

Author	Sample	Methodology	Findings
Brander et al. (2019)	4,085,367 individuals, including 22,232 OCD probands	Family study	Familial clustering of OCD, especially tic-related OCD. Using hazard ratios adjusted for sex and birth year, the risk of OCD in full siblings of patients with tic-related OCD and patients with non-tic-related OCD was estimated at 10.63 and 4.52, respectively.
Burton et al. (2018)	220 pairs of non-clinical twins	Twin study	Heritability of obsessive-compulsive traits was estimated at 74%. Shared genetic factors explained most of the variance shared among the obsessive-compulsive trait dimensions examined in this study, which included cleaning/contamination, symmetry/ordering, rumination, superstition, counting/checking, and hoarding.
Stewart et al. (2013)	1,465 OCD probands and 5,557 control subjects	Genome-wide association study (GWAS)	No genome-wide significant single nucleotide polymorphism (SNP) was detected in case- control analysis, which yielded the top SNPs located within DLGAP1. Trio-based analysis revealed a SNP reaching genome-wide significance located near BTBD3. Combining the two analyses revealed significant enrichment for regions regulating gene expression and methylation in the brain.
Mattheisen et al. (2015)	1,406 OCD probands and 5,061 control subjects	GWAS	No genome-wide significant SNP was detected. Top-ranked SNPs were near PTPRD, CDH9, and CDH10. Two SNPs near GRIK2 and HACE1 detected in the first OCD GWAS were also among the significant SNPs.
Den Braber et al. (2016)	Obsessive-compulsive symptoms of 8,267 subjects	GWAS	A genome-wide significant SNP was detected at MEF2BNB. Gene-based testing revealed four genes significantly associated with obsessive-compulsive symptoms located in the 19p13.11 chromosomal region, which has been associated with brain and immune processes.
Costas et al. (2016)	370 OCD probands and 443 control subjects	GWAS	No genome-wide significant SNP was detected. A polygenic risk score (PRS) estimated using the Psychiatric Genetics Consortium (PGC) schizophrenia (SCZ) meta-analysis sample significantly predicted OCD case-control status among the study sample.
Guo et al. (2017)	2,998 OCD case-control samples and 6,898 autism spectrum disorder (ASD) case-control samples	GWAS	No genome-wide significant SNP was detected. Top-ranked SNP identified is in linkage disequilibrium (LD) with TUBB3. A PRS estimated using the ASD case-control sample significantly predicted case-control status in the OCD sample.
PGC OCD workgroup (2018)	2,688 OCD probands and 7,037 control subjects	GWAS	No genome-wide significant SNP was detected. Most significant SNPs were in LD with GRID2 and KIT. PRSs estimated based on the first and second OCD GWASs significantly predicted case-control status in the second and the first OCD GWASs case-control samples, respectively. SNP-based heritability was estimated at 0.28.
Khramtsova et al. (2018)	PGC-OCD GWAS sample	GWAS	SNP-based heritability estimates were similar for male and female OCD, and significant genetic correlation between them was reported. GRID2 and GRP135 were associated only with female OCD in gene-based tests. SNPs with sex-specific effects were significantly enriched for regions regulating gene expression in brain and immune tissues.
Smit et al. (2020)	PGC-OCD GWAS sample and compulsive symptoms of 8,267 subjects	GWAS	No genome-wide significant SNP was detected. Significant genetic correlation between OCD and compulsive symptoms was found. WDR7, ADCK1, GRID2, and KIT were detected in gene-based tests. Top genes detected in gene-based tests were significantly enriched for genes expressed in the anterior cingulate cortex, nucleus accumbens, and amygdala.
Cross-Disorder Group of the PGC (2019)	232,964 probands diagnosed with anorexia nervosa (AN), attention deficit hyperactivity disorder (ADHD), ASD, bipolar disorder (BD), major depressive disorder (MDD), OCD, SCZ, or Tourette’s syndrome (TS) and 494,162 controls subjects	GWAS	Genetic correlation between OCD and AN was among the highest observed. Exploratory factor analysis suggested shared genetic liability among OCD, AN, and TS.
Bralten et al. (2020)	650 non-clinical children and adolescents	GWAS	The individual results from the GWAS of obsessive-compulsive symptoms in the sample of 650 non-clinical children and adolescents were not reported owing to insufficient power. Nonetheless, PRS analysis revealed shared genetic etiology among OCD, obsessive- compulsive symptoms, and insulin signaling-related traits.
Cappi et al. (2016)	17 OCD parent–child trios	Whole-exome sequencing study	20 nonsynonymous de novo single nucleotide variants (SNVs) detected among 17 OCD probands. SNVs were enriched in pathways related to the immune system and neurodevelopment.
Cappi et al. (2019)	184 OCD and 777 control parent– child trios	Whole-exome sequencing study	A significantly higher prevalence of de novo damaging variants (33.9%) was found among probands. These variants were estimated to be present in 22.2% of overall OCD cases. SCUBE1 and CHD8 were identified as high-confidence risk genes for OCD. Genes carrying these variants significantly overlap among ASD and TS probands and were associated with immune system-related processes.
Gazzellone et al. (2016)	307 pediatric OCD probands and 3,861 control subjects	Copy number variants study	A similar rate of rare copy number variants (CNVs) in early onset OCD probands and controls was observed. In 5.9% of probands, structural variants were detected in genes associated with brain function. Particularly, CNVs were found in three targets of fragile X mental retardation protein (FMRP), including DLGAP1 and PTPRD.
Grünblatt et al. (2017)	121 pediatric OCD probands and 123 control subjects	Copy number variants study	Although no significant difference in number and size of rare CNVs between pediatric OCD patients and controls was detected, the number of CNVs encompassing genes involved in neurological function was higher among the former.
Zarrei et al. (2019)	2,691 probands diagnosed with OCD, SCZ, ASD, or ADHD and 1,769 family members	Copy number variants study	Clinically relevant CNVs were found in 5.6% of OCD probands. Multiple brain-expressed genes impacted by CNVs in at least two probands were found. An increased burden of rare CNVs impacting genes involved in genomic stability was found in probands.
Wang et al. (2018)	77 OCD, 39 MDD, and 40 SCZ probands and 67 control subjects	Gene expression analysis	A blood-based gene expression panel (including FKBP1A, COPS7A, FIBP, TP73-AS1, SDF4, and GOLGA8A) was able to distinguish OCD probands from MDD and SCZ probands and healthy controls with 88% sensitivity and 85% specificity.
Piantadosi et al. (2019)	Postmortem brain samples of 8 OCD probands and 8 control subjects	Gene expression analysis	Overall reduced expression of genes related to excitatory synapses in the orbitofrontal cortex (OFC), but not in striatal regions, of OCD probands. Minor alterations were observed in the expression of genes related to inhibitory synapses in the OFC of OCD probands.
Nissen et al. (2016)	21 pediatric and adolescent OCD probands and 12 female controls	DNA methylation analysis	No significantly different DNA methylation levels in candidate genes previously implicated in OCD were found between probands and controls.
Yu et al. (2016)	65 pediatric OCD probands and 96 control subjects	DNA methylation analysis	Multiple genes with different DNA methylation levels among OCD cases and controls were found. Further enrichment analysis implicated those genes in cell adhesion- and actin-related processes.
Alemany-Navarro et al. (2019)	103 OCD probands	Gene–environment interaction	Although not predicting treatment response, a PRS estimate based on OCD risk variants predicted pretreatment symptom severity.
Mahjani et al. (2020)	822,843 individuals, including 7,184 probands diagnosed with OCD	Maternal effects	Genetic maternal effects (i.e. the influence of the maternal genotype on the phenotype of the offspring) accounted for 7.6% of the liability for OCD, whereas additive genetic effects accounted for 35%.
Trankner et al. (2019)		Experimental animal system	Targeted deletion of Hoxb8-lineage microglia led to significant grooming and anxiety- like behaviors and stress-response among female rodents, which were ameliorated by suppressing female sex hormones.
Delgado- Acevedo et al. (2018)		Experimental animal system	Overexpression of EAAT3 in the frontal cortex, hippocampus, and striatum increased anxious and grooming behaviors and prolonged spontaneous recovery of conditioned fear among rodents. Moreover, alterations in NMDA receptor constitution and corticostriatal synaptic plasticity were observed.
Zike et al. (2017)		Experimental animal system	Constitutive reduced expression of EAAT3 in rodents caused dampened stereotyped response following dopaminergic challenges, which was rescued by restoration of EAAT3 expression in the midbrain.
Krabbe et al. (2017)		Experimental animal system	Rodent with microglia-restricted progranulin (PGRN) inactivation exhibited increased self- grooming and marble-burying behaviors, which were normalized with suppression of nuclear factor κB (NF-κB) signaling.
Ullrich et al. (2018)		Experimental animal system	SPRED2 knockout mice displayed severe grooming and anxiety behaviors, dysfunctional thalamo-amygdala synapses, and altered expression of synaptic proteins in the amygdala. Inhibition of overactive (TrkB)/ERK-MAPK signaling ameliorated pathologic behavior.
Van de Vondervoort et al. (2020)		Experimental animal system	TALLYHO/JngJ rodents, which are a model of human type 2 diabetes mellitus, exhibited increased compulsive and anxious behaviors in addition to structural brain abnormalities in midline corpus callosum, dorsomedial striatum, and superior cerebellar peduncles.
Noh et al. (2017)	592 OCD probands and 560 control subjects	Functional variants prioritization	Analysis of 608 genes potentially associated with OCD in human, rodent, and dog studies revealed significant functional variant burden in four genes (NRXN1, HTR2A, CTTNBP2, and REEP3) involved in brain pathways implicated in OCD. NRXN1 achieved genome-wide significance after inclusion of GWAS data from 33,370 controls.
Rodriguez et al. (2017)	102 OCD early onset OCD probands and 47 control subjects	Human cell models	Higher percentages of total monocytes and CD16+ monocytes in OCD probands as compared to control subjects. Increased production of proinflammatory cytokines by proband monocytes after stimulation with lipopolysaccharide.
Özyurt et al. (2019)	60 drug-naïve adolescent OCD probands and 128 adolescent control subjects	Human cell models	Significant differences were detected in the neutrophil: lymphocyte ratio and white blood cell, neutrophil, and platelet counts among OCD probands with comorbid anxiety disorder, OCD probands with no comorbidities, and control subjects, which remained after controlling for age and sex.
Hibar et al. (2018)	First OCD GWAS sample and neuroimaging GWASs of volume of eight subcortical brain regions (13,171 subjects)	Imaging genetics	OCD risk SNPs were associated with increased volumes of the nucleus accumbens and the putamen. Conditional false discovery rate analysis revealed specific SNPs influencing OCD risk and putamen, amygdala, and thalamus volumes.
Zartaloudi et al. (2019)	707 unaffected relatives of OCD probands and 842 control subjects	Meta-analysis of endophenotype studies	21 studies were included in the meta-analysis. A significant impairment in global executive functioning was found among unaffected relatives of OCD probands, with specific deficits in planning, visuospatial working memory, and verbal fluency.
Qin et al. (2016)	804 OCD probands (514 responders and 290 non-responders)	Pharmacogenetics	A genome-wide significant SNP was detected near the gene DISP1, within a chromosomal region associated with neurodevelopment. Most significant SNPs were enriched for biological pathways related to glutamate and serotonin neurotransmission.
Lisoway et al. (2018)	112 OCD probands	Pharmacogenetics	No significant associations between a SNP of DISP1 and response to serotonin reuptake inhibitor treatment.