Converting single nucleotide variants between genome builds: from cautionary tale to solution.

Next-generation sequencing studies are dependent on a high-quality reference genome for single nucleotide variant (SNV) calling. Although the two most recent builds of the human genome are widely used, position information is typically not directly comparable between them. Re-alignment gives the most accurate position information, but this procedure is often computationally expensive, and therefore, tools such as liftOver and CrossMap are used to convert data from one build to another. However, the positions of converted SNVs do not always match SNVs derived from aligned data, and in some instances, SNVs are known to change chromosome when converted. This is a significant problem when compiling sequencing resources or comparing results across studies. Here, we describe a novel algorithm to identify positions that are unstable when converting between human genome reference builds. These positions are detected independent of the conversion tools and are determined by the chain files, which provide a mapping of contiguous positions from one build to another. We also provide the list of unstable positions for converting between the two most commonly used builds GRCh37 and GRCh38. Pre-excluding SNVs at these positions, prior to conversion, results in SNVs that are stable to conversion. This simple procedure gives the same final list of stable SNVs as applying the algorithm and subsequently removing variants at unstable positions. This work highlights the care that must be taken when converting SNVs between genome builds and provides a simple method for ensuring higher confidence converted data. Unstable positions and algorithm code, available at https://github.com/cathaloruaidh/genomeBuildConversion.

Gender bias in academic medicine: a resumé study.

BACKGROUND: Minimising the effects of unconscious bias in selection for clinical academic training is essential to ensure that allocation of training posts is based on merit. We looked at the effect of anonymising applications to a training programme for junior doctors on the scores of the applications and on gender balance; and whether female candidates were more likely to seek gender-concordant mentors. METHODS: Applications to the training programme were reviewed and scored independently by reviewers who received either an anonymised or named copy. Scores were compared using a paired t-test, and differences in scores compared by gender. The gender of named supervisors for male and female candidates was compared. RESULTS: Scores of 101 applications were reviewed. When their identity was known, male candidates scored 1.72% higher and female candidates scored 0.74% higher, but these findings were not statistically significant (p value = 0.279 and 0.579). Following introduction of anonymisation, the proportion of successful female candidates increased from 27 to 46%. Female candidates were more likely to name a female supervisor compared to male (41% vs. 25% of supervisors). CONCLUSIONS: Anonymising applications did not significantly change scores, although gender balance improved. Gender-concordant mentoring initiatives should consider effects on mentors as well as mentees.

Identifying schizophrenia patients who carry pathogenic genetic copy number variants using standard clinical assessment: retrospective cohort study.

BACKGROUND: Copy number variants (CNVs) play a significant role in disease pathogenesis in a small subset of individuals with schizophrenia (~2.5%). Chromosomal microarray testing is a first-tier genetic test for many neurodevelopmental disorders. Similar testing could be useful in schizophrenia. AIMS: To determine whether clinically identifiable phenotypic features could be used to successfully model schizophrenia-associated (SCZ-associated) CNV carrier status in a large schizophrenia cohort. METHOD: Logistic regression and receiver operating characteristic (ROC) curves tested the accuracy of readily identifiable phenotypic features in modelling SCZ-associated CNV status in a discovery data-set of 1215 individuals with psychosis. A replication analysis was undertaken in a second psychosis data-set (n = 479). RESULTS: In the discovery cohort, specific learning disorder (OR = 8.12; 95% CI 1.16-34.88, P = 0.012), developmental delay (OR = 5.19; 95% CI 1.58-14.76, P = 0.003) and comorbid neurodevelopmental disorder (OR = 5.87; 95% CI 1.28-19.69, P = 0.009) were significant independent variables in modelling positive carrier status for a SCZ-associated CNV, with an area under the ROC (AUROC) of 74.2% (95% CI 61.9-86.4%). A model constructed from the discovery cohort including developmental delay and comorbid neurodevelopmental disorder variables resulted in an AUROC of 83% (95% CI 52.0-100.0%) for the replication cohort. CONCLUSIONS: These findings suggest that careful clinical history taking to document specific neurodevelopmental features may be informative in screening for individuals with schizophrenia who are at higher risk of carrying known SCZ-associated CNVs. Identification of genomic disorders in these individuals is likely to have clinical benefits similar to those demonstrated for other neurodevelopmental disorders.

Review of statistical methodologies for the detection of parent-of-origin effects in family trio genome-wide association data with binary disease traits.

The detection of parent-of-origin effects aims to identify whether the functionality of alleles, and in turn associated phenotypic traits, depends on the parental origin of the alleles. Different parent-of-origin effects have been identified through a variety of mechanisms and a number of statistical methodologies for their detection have been proposed, in particular for genome-wide association studies (GWAS). GWAS have had limited success in explaining the heritability of many complex disorders and traits, but successful identification of parent-of-origin effects using trio (mother, father and offspring) GWAS may help shed light on this missing heritability. However, it is important to choose the most appropriate parent-of-origin test or methodology, given knowledge of the phenotype, amount of available data and the type of parent-of-origin effect(s) being considered. This review brings together the parent-of-origin detection methodologies available, comparing them in terms of power and type I error for a number of different simulated data scenarios, and finally offering guidance as to the most appropriate choice for the different scenarios.

Traumatic Brain Injury, Sleep, and Mental Health: A Longitudinal Study of Air Force Personnel Pre- and Postdeployment to Iraq.

OBJECTIVE: We investigated the complex relationships between traumatic brain injury (TBI), sleep, and mental health problems longitudinally among US service members (SMs) pre- and postdeployment to Iraq. PARTICIPANTS: One hundred sixty-eight SMs enrolled in a 4-week Air Force Basic Combat Convoy Course predeployment. DESIGN: Self-report data were collected at the beginning and end of training and then at 1, 3, 6, and 12 months postdeployment. Regression analyses were implemented, and participants were categorized into 4 groups based on TBI history for further statistical analysis. RESULTS: Positive TBI history was associated with greater symptoms of insomnia and posttraumatic stress predeployment and persistence of insomnia symptoms, posttraumatic stress, and depression postdeployment. Positive TBI history and posttraumatic stress served as risk factors for head injury in Iraq, and SMs who reported a head injury during deployment also endorsed greater posttraumatic stress postdeployment than those without head injury. SMs with positive TBI history who also reported a new TBI in Iraq endorsed the greatest sleep and mental health problems across the study period. CONCLUSIONS: This study provides valuable information regarding temporal relationships between TBI, sleep, and mental health problems among a combat military population. Findings have important implications from both prevention and clinical perspectives.

Belonging protects against postdeployment depression in military personnel.

BACKGROUND: Depression among U.S. military personnel has received relatively little empirical attention compared to posttraumatic stress disorder, despite evidence that depression is associated with poor psychosocial outcomes and increased suicide risk. Even less is known about factors that protect against depression in military populations. METHODS: A sample of 168 active duty Air Force convoy operators completed self-report measures of depression, posttraumatic stress, and sense of "belonging" before deploying to Iraq, and again at 1, 3, 6, and 12 months following their return. Linear growth modeling was used to test the associations of the variables over time. RESULTS: Mean depression scores remained low and stable across the deployment and 12-month follow-up period. Increased depression severity was significantly associated with low belonging (P < .001) and with posttraumatic stress symptoms (P < .001) at every time point. LIMITATIONS: Relatively small, predominantly male sample utilizing self-report methods. CONCLUSIONS: A sense of belongingness may protect service members from depression at all stages of the deployment cycle, from predeployment preparations through deployment and postdeployment adjustment.

DEPRESSION MEDIATES THE RELATION OF INSOMNIA SEVERITY WITH SUICIDE RISK IN THREE CLINICAL SAMPLES OF U.S. MILITARY PERSONNEL.

BACKGROUND: A growing body of empirical research suggests insomnia severity is directly related to suicide ideation, attempts, and death in nonmilitary samples, even when controlling for depression and other suicide risk factors. Few studies have explored this relationship in U.S. military personnel. METHODS: The present study entailed secondary data analyses examining the associations of insomnia severity with suicide ideation and attempts in three clinical samples: Air Force psychiatric outpatients (n = 158), recently discharged Army psychiatric inpatients (n = 168), and Army psychiatric outpatients (n = 54). Participants completed the Beck Scale for Suicide Ideation, the Beck Depression Inventory-II or Patient Health Questionnaire-9, the Insomnia Severity Index, and the Posttraumatic Stress Disorder Checklist at baseline; two samples also completed these measures during follow-up. RESULTS: Sleep disturbance was associated with concurrent (ß's > 0.21; P's < 0.059) and prospective (ß's > 0.39; P's < 0.001) suicide ideation in all three samples. When adjusting for age, gender, depression, and posttraumatic stress, insomnia severity was no longer directly associated with suicide ideation either concurrently (ß's < 0.19; P's > 0.200) or prospectively (ß's < 0.26; P's > 0.063), but depression was (ß's > 0.22; P's < 0.012). Results of a latent difference score mediation model indicated that depression mediated the relation of insomnia severity with suicide ideation. CONCLUSIONS: Across three clinical samples of military personnel, depression explained the relationship between insomnia severity and suicide risk.

Individual common variants exert weak effects on the risk for autism spectrum disorders.

While it is apparent that rare variation can play an important role in the genetic architecture of autism spectrum disorders (ASDs), the contribution of common variation to the risk of developing ASD is less clear. To produce a more comprehensive picture, we report Stage 2 of the Autism Genome Project genome-wide association study, adding 1301 ASD families and bringing the total to 2705 families analysed (Stages 1 and 2). In addition to evaluating the association of individual single nucleotide polymorphisms (SNPs), we also sought evidence that common variants, en masse, might affect the risk. Despite genotyping over a million SNPs covering the genome, no single SNP shows significant association with ASD or selected phenotypes at a genome-wide level. The SNP that achieves the smallest P-value from secondary analyses is rs1718101. It falls in CNTNAP2, a gene previously implicated in susceptibility for ASD. This SNP also shows modest association with age of word/phrase acquisition in ASD subjects, of interest because features of language development are also associated with other variation in CNTNAP2. In contrast, allele scores derived from the transmission of common alleles to Stage 1 cases significantly predict case status in the independent Stage 2 sample. Despite being significant, the variance explained by these allele scores was small (Vm< 1%). Based on results from individual SNPs and their en masse effect on risk, as inferred from the allele score results, it is reasonable to conclude that common variants affect the risk for ASD but their individual effects are modest.

Investigating copy number variants in schizophrenia pedigrees using a new consensus pipeline called PECAN.

Copy number variants (CNVs) have been implicated in many human diseases, including psychiatric disorders. Whole genome sequencing offers advantages in CNV calling compared to previous array-based methods. Here we present a robust and transparent CNV calling pipeline, PECAN (PEdigree Copy number vAriaNt calling), for short-read, whole genome sequencing data, comprised of a novel combination of four calling methods and structural variant genotyping. This method is scalable and can incorporate pedigree information to retain lower-confidence CNVs that would otherwise be discarded. We have robustly benchmarked PECAN using gold-standard CNV calls for two well-established evaluation samples, NA12878 and HG002, showing that PECAN performs with high precision and recall on both datasets, outperforming another pedigree-based CNV calling pipeline. As part of this work, we provide a list of high-confidence gold standard CNVs for the NA12878 reference sample, curated from multiple studies. We applied PECAN to a collection of pedigrees multiply affected with schizophrenia and identified a rare deletion that perfectly co-segregates with schizophrenia in one of the pedigrees. The CNV overlaps the gene PITRM1, which has been implicated in a complex phenotype including ataxia, developmental delay, and schizophrenia-like episodes in affected adults.

Whole genome sequencing study of identical twins discordant for psychosis.

Monozygotic (MZ) twins are often thought to have identical genomes, but recent work has shown that early post-zygotic events can result in a spectrum of DNA variants that are different between MZ twins. Such variants may explain phenotypic discordance and contribute to disease etiology. Here we performed whole genome sequencing in 17 pairs of MZ twins discordant for a psychotic disorder (schizophrenia, schizoaffective disorder or bipolar disorder). We examined various classes of rare variants that are discordant within a twin pair. We identified four genes harboring rare, predicted deleterious missense variants that were private to an affected individual in the cohort. Variants in FOXN1 and FLOT2 would have been categorized as damaging from recent schizophrenia and bipolar exome sequencing studies. Additionally, we identified four rare genic copy number variants (CNVs) private to an affected sample, two of which overlapped genes that have shown evidence for association with schizophrenia or bipolar disorder. One such CNV was a 3q29 duplication previously implicated in autism and developmental delay. We have performed the largest MZ twin study for discordant psychotic phenotypes to date. These findings warrant further investigation using other analytical approaches.

P53 and TLR4 expression are prognostic markers informing progression free survival of advanced stage high grade serous ovarian cancer.

OBJECTIVE: New prognostic biomarkers, and bio-signatures, are urgently needed to facilitate a precision medicine-based approach to more effectively treat patients with high-grade serous ovarian cancer (HGSC). In this study, we analysed the expression patterns of a series of candidate protein biomarkers. METHODS: The panel of markers which included MyD88, TLR4, MAD2, PR, OR, WT1, p53, p16, CD10 and Ki67 was assessed using immunohistochemistry in a tissue microarray (TMA) cohort of n = 80 patients, composed of stage 3-4 HGSCs. Each marker was analysed for their potential to predict both overall survival (OS) and progression-free survival (PFS). RESULTS: TLR4 and p53 were found to be individually predictive of poorer PFS (Log Rank, p = 0.017, p = 0.030 respectively). Cox regression analysis also identified high p53 and TLR4 expression as prognostic factors for reduced PFS (p53; HR=1.785, CI=1.036-3.074, p = 0.037 and TLR4; HR=2.175, CI=1.112-4.253, p = 0.023). Multivariate forward conditional Cox regression analysis, examining all markers, identified a combined signature composed of p53 and TLR4 as prognostic for reduced PFS (p = 0.023). CONCLUSION: Combined p53 and TLR4 marker assessment may help to aid treatment stratification for patients diagnosed with advanced-stage HGSC.

Military mental health: the role of daily hassles while deployed.

This study sought to identify factors contributing to symptoms of depression and posttraumatic stress disorder (PTSD) in recently deployed combat veterans. A sample of 168 active duty military personnel completed measures of combat exposure, deployment-related daily hassles, depression symptoms, and PTSD symptoms at six time points across their deployment: predeployment and 1, 3, 6, and 12 months postdeployment. Mixed-effects linear modeling with repeated measures was used to identify factors associated with depression and PTSD severity over time. Postdeployment depression severity did not change over time, but PTSD severity decreased slightly over time after returning home. Postdeployment depression severity was predicted by past (but not recent) combat exposure, daily hassles, and concurrent PTSD symptoms. Postdeployment PTSD severity was predicted by past and recent combat exposure, concurrent depression symptoms, and male sex. Depression severity mediated the relationship between daily hassles and postdeployment PTSD severity.

Evidence for parent-of-origin effects in autism spectrum disorder: a narrative review.

Autism spectrum disorder (ASD) is a heterogeneous group of early-onset neurodevelopmental disorders known to be highly heritable with a complex genetic architecture. Abnormal brain developmental trajectories that impact synaptic functioning, excitation-inhibition balance and brain connectivity are now understood to play a central role in ASD. Ongoing efforts to identify the genetic underpinnings still prove challenging, in part due to phenotypic and genetic heterogeneity.This review focuses on parent-of-origin effects (POEs), where the phenotypic effect of an allele depends on its parental origin. POEs include genomic imprinting, transgenerational effects, mitochondrial DNA, sex chromosomes and mutational transmission bias. The motivation for investigating these mechanisms in ASD has been driven by their known impacts on early brain development and brain functioning, in particular for the most well-documented POE, genomic imprinting. Moreover, imprinting is implicated in syndromes such as Angelman and Prader-Willi, which frequently share comorbid symptoms with ASD. In addition to other regions in the genome, this comprehensive review highlights the 15q11-q13 and 7q chromosomal regions as well as the mitochondrial DNA as harbouring the majority of currently identified POEs in ASD.

Ultrarare Missense Variants Implicated in Utah Pedigrees Multiply Affected With Schizophrenia.

Background: Recent work from the Schizophrenia Exome Sequencing Meta-analysis (SCHEMA) consortium showed significant enrichment of ultrarare variants in schizophrenia cases. Family-based studies offer a unique opportunity to evaluate rare variants because risk in multiplex pedigrees is more likely to be influenced by the same collection of variants than an unrelated cohort. Methods: Here, we examine whole genome sequencing data from 35 individuals across 6 pedigrees multiply affected by schizophrenia. We applied a rigorous filtering pipeline to search for classes of protein-coding variants that cosegregated with disease status, and we examined these for evidence of enrichment in the SCHEMA dataset. Additionally, we applied a family-based consensus approach to call copy number variants and screen against a list of schizophrenia-associated risk variants. Results: We identified deleterious missense variants in 3 genes (ATP2B2, SLC25A28, and GSK3A) that cosegregated with disease in 3 of the pedigrees. In the SCHEMA, the gene ATP2B2 shows highly suggestive evidence for deleterious missense variants in schizophrenia cases (p = .000072). ATP2B2 is involved in intracellular calcium homeostasis, expressed in multiple brain tissue types, and predicted to be intolerant to loss-of-function and missense variants. Conclusions: We have identified genes that are likely to increase schizophrenia risk in 3 of the 6 pedigrees examined, the strongest evidence being for a gene involved in calcium homeostasis. Further work is required to examine other classes of variants that may be contributing to disease burden.

A genome-wide scan for common alleles affecting risk for autism.

Although autism spectrum disorders (ASDs) have a substantial genetic basis, most of the known genetic risk has been traced to rare variants, principally copy number variants (CNVs). To identify common risk variation, the Autism Genome Project (AGP) Consortium genotyped 1558 rigorously defined ASD families for 1 million single-nucleotide polymorphisms (SNPs) and analyzed these SNP genotypes for association with ASD. In one of four primary association analyses, the association signal for marker rs4141463, located within MACROD2, crossed the genome-wide association significance threshold of P < 5 × 10(-8). When a smaller replication sample was analyzed, the risk allele at rs4141463 was again over-transmitted; yet, consistent with the winner's curse, its effect size in the replication sample was much smaller; and, for the combined samples, the association signal barely fell below the P < 5 × 10(-8) threshold. Exploratory analyses of phenotypic subtypes yielded no significant associations after correction for multiple testing. They did, however, yield strong signals within several genes, KIAA0564, PLD5, POU6F2, ST8SIA2 and TAF1C.

Exploration of empirical Bayes hierarchical modeling for the analysis of genome-wide association study data.

In the analysis of genome-wide association (GWA) data, the aim is to detect statistical associations between single nucleotide polymorphisms (SNPs) and the disease or trait of interest. These SNPs, or the particular regions of the genome they implicate, are then considered for further study. We demonstrate through a comprehensive simulation study that the inclusion of additional, biologically relevant information through a 2-level empirical Bayes hierachical model framework offers a more robust method of detecting associated SNPs. The empirical Bayes approach is an objective means of analyzing the data without the need for the setting of subjective parameter estimates. This framework gives more stable estimates of effects through a reduction of the variability in the usual effect estimates. We also demonstrate the consequences of including additional information that is not informative and examine power and false-positive rates. We apply the methodology to a number of genome-wide association (GWA) data sets with the inclusion of additional biological information. Our results agree with previous findings and in the case of one data set (Crohn's disease) suggest an additional region of interest.

Service user satisfaction with care in a specialist service for young people with attention deficit hyperactivity disorder.

OBJECTIVES: Consumer satisfaction is considered one of the most important measures of service quality in child mental health; however, there is limited understanding of factors that influence satisfaction. The objective of this study was to investigate key factors influencing satisfaction with care (SWC) in ADMiRE, a specialist service for young people (YP) with attention deficit hyperactivity disorder (ADHD). METHODS: Parents/carers (n = 67) and YP > 9 years (n = 44) attending ADMiRE completed an anonymous Experience of Service Questionnaire (ESQ), a quantitative/qualitative measure of service user satisfaction. Parents/carers also completed symptom severity rating scales. Data were analysed to determine (i) overall SWC, (ii) the relationship between parent- and youth-reported SWC and (iii) the impact of symptom severity on SWC. Thematic analysis of qualitative ESQ data was completed. RESULTS: Parents/carers were significantly more satisfied than YP (p = 0.028). Symptom severity did not impact significantly on parent/carer satisfaction. YP with severe hyperactive/impulsive and inattentive ADHD symptoms were significantly less satisfied with care than those with less severe ADHD symptoms (p = 0.022 and p = 0.017 respectively). Factors related to the therapeutic alliance were identified as being particularly important to both parents/carers and YP. CONCLUSIONS: This is the first Irish study that has investigated the impact of symptom severity on service user satisfaction in a child mental health service. The results highlight the different perspectives of YP and parents and provide novel insights into the impact of symptom severity on service user satisfaction. The importance of the therapeutic alliance should not be underestimated in future development of services.

The SNP ratio test: pathway analysis of genome-wide association datasets.

UNLABELLED: We present a tool that assesses the enrichment of significant associations from genome-wide association studies (GWAS) in a pathway context. The SNP ratio test (SRT) compares the proportion of significant to all SNPs within genes that are part of a pathway and computes an empirical P-value based on comparisons to ratios in datasets where the assignment of case/control status has been randomized. We applied the SRT to a Parkinson's disease GWAS dataset, using the KEGG database, revealing significance for Parkinson's disease and related pathways. AVAILABILITY: https://sourceforge.net/projects/snpratiotest/

Outcome of First-admission Depression Treated in a Specialized Mood Disorders Service.

OBJECTIVE: Few studies have described the treatment or outcome of depression in specialized mood disorders units (MDUs). Previous studies have focused on cohorts of patients with highly treatment-resistant illness who are likely to have a poor prognosis even with intensive treatment. This study describes the treatment and medium-term outcomes of a cohort of first-admission depressed patients with less treatment-resistant illness treated in a specialized MDU. METHODS: A cohort of 137 consecutive first-admission depressed patients, referred to an MDU over 2 years, were interviewed using standardized schedules and followed up prospectively from admission for ∼18 months to describe baseline characteristics, treatment, outcome, and predictors of outcome. Times to recovery and recurrence were evaluated using survival analyses and predictors of outcome were examined using bivariate and multivariate regression analyses. RESULTS: On admission, 75% of the 137 patients had depression that had been found to be resistant to pharmacological treatment, and 34% had been chronically depressed (>2 y). Over half of the patients had likely maladaptive personality traits and one third had at least 1 comorbid psychiatric disorder. By discharge, a significantly higher proportion of the patients were being prescribed very high (P<0.01) or high doses (P<0.05) of antidepressants, augmentation therapy (P<0.001), or a combination of antidepressants (P<0.001) or were engaged in individual psychotherapy (P<0.001), compared with baseline. With intensive treatment, 62% of the patients recovered by 6 months and 76% by 12 months, with 83% overall recovering and patients found to be asymptomatic during almost 60% of the follow-up period. However, 48% suffered a recurrence over the course of the follow-up. Chronicity of mood episodes (P<0.01) and the presence of psychiatric comorbidity (P<0.05) predicted recurrence. CONCLUSIONS: This prospective, naturalistic, medium-term study describes better outcomes, in terms of recovery and symptomatology over time, in a cohort of first-admission depressed patients than previous first-admission studies after continuous, intensive treatment, although the proportion of patients who experienced recurrences remained high.

Prevalence of N-Methyl-d-Aspartate Receptor antibody (NMDAR-Ab) encephalitis in patients with first episode psychosis and treatment resistant schizophrenia on clozapine, a population based study.

INTRODUCTION: N-methyl-d-aspartate receptor antibody (NMDAR-Ab) encephalitis consensus criteria has recently been defined. We aimed to examine the prevalence of NMDAR-Ab encephalitis in patients with first episode psychosis (FEP) and treatment resistant schizophrenia (TRS) on clozapine, using clinical investigations, antibody testing and to retrospectively apply diagnostic consensus criteria. METHODS: Adult (18-65 years old) cases of FEP meeting inclusion criteria were recruited over three years and assessed using the Structured Clinical Interview for DSM-IV disorders (SCID). NMDAR-Ab was identified using a live cell-based assay (L-CBA). Seropositive cases were clinically investigated for features of encephalitis including neuro-imaging, EEG and CSF where possible. Serum was retested using immunohistochemistry (IHC) as part of diagnostic criteria guidelines. A cohort of patients with TRS was also recruited. RESULTS: 112 FEP patients were recruited over 3 years. NMDAR-Ab seroprevalence was 4/112 (3.5%) cases. One case (<1%) was diagnosed with definite NMDAR-Ab encephalitis and treated with immunotherapy. One of the three other seropositive cases met criteria for probable encephalitis. However all three were ultimately diagnosed with mood disorders with psychotic features. None have developed neurological features at three year follow up. 1/100 (1%) of patients with TRS was 100 patients with TRS were recruited. One case (1%) seropositive for NMDAR-Ab but did not meet criteria for encephalitis. CONCLUSIONS: NMDAR-Ab encephalitis as defined by consensus guidelines occured rarely in psychiatric services in this study. Further studies are needed to establish pathogenicity of serum NMDAR-Ab antibodies. Psychiatric services should be aware of the clinical features of encephalitis.