Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Open Access

Peer-reviewed

Research Article

Primary complex motor stereotypies are associated with de novo damaging DNA coding mutations that identify KDM5B as a risk gene

  • Thomas V. Fernandez ,

    Roles Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing

    thomas.fernandez@yale.edu

    Affiliations Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America, Department of Psychiatry, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Zsanett P. Williams,

    Roles Data curation, Formal analysis, Project administration, Writing – review & editing

    Affiliation Department of Psychiatry, Vanderbilt University School of Nursing, Nashville, TN, United States America

    ⨯
  • Tina Kline,

    Roles Data curation, Investigation, Methodology, Project administration, Writing – review & editing

    Affiliation Departments of Neurology and Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, United States America

    ⨯
  • Shreenath Rajendran,

    Roles Data curation, Methodology, Project administration, Validation, Writing – review & editing

    Affiliation Departments of Neurology and Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, United States America

    ⨯
  • Farhan Augustine,

    Roles Data curation, Investigation, Methodology, Project administration, Validation

    Affiliation Departments of Neurology and Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, United States America

    ⨯
  • Nicole Wright,

    Roles Data curation, Investigation, Methodology, Project administration, Validation, Writing – review & editing

    Affiliation Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Catherine A. W. Sullivan,

    Roles Data curation, Investigation, Methodology, Validation, Writing – review & editing

    Affiliation Department of Pediatrics, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Emily Olfson,

    Roles Investigation, Methodology, Writing – review & editing

    Affiliation Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Sarah B. Abdallah,

    Roles Investigation, Methodology, Writing – review & editing

    Affiliation Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Wenzhong Liu,

    Roles Data curation, Investigation, Methodology, Project administration, Validation, Writing – review & editing

    Affiliation Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Ellen J. Hoffman,

    Roles Conceptualization, Investigation, Methodology, Writing – review & editing

    Affiliation Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Abha R. Gupta,

    Roles Conceptualization, Data curation, Formal analysis, Investigation, Validation, Writing – original draft, Writing – review & editing

    Affiliations Yale Child Study Center, Yale University School of Medicine, New Haven, CT, United States America, Department of Pediatrics, Yale University School of Medicine, New Haven, CT, United States America

    ⨯
  • Harvey S. Singer

    Roles Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Resources, Supervision, Validation, Writing – original draft, Writing – review & editing

    Affiliation Departments of Neurology and Pediatrics, Johns Hopkins University School of Medicine, Baltimore, MD, United States America

    ⨯

Abstract

Motor stereotypies are common in children with autism spectrum disorder (ASD), intellectual disability, or sensory deprivation, as well as in typically developing children (“primary” stereotypies, pCMS). The precise pathophysiological mechanism for motor stereotypies is unknown, although genetic etiologies have been suggested. In this study, we perform whole-exome DNA sequencing in 129 parent-child trios with pCMS and 853 control trios (118 cases and 750 controls after quality control). We report an increased rate of de novo predicted-damaging DNA coding variants in pCMS versus controls, identifying KDM5B as a high-confidence risk gene and estimating 184 genes conferring risk. Genes harboring de novo damaging variants in pCMS probands show significant overlap with those in Tourette syndrome, ASD, and those in ASD probands with high versus low stereotypy scores. An exploratory analysis of these pCMS gene expression patterns finds clustering within the cortex and striatum during early mid-fetal development. Exploratory gene ontology and network analyses highlight functional convergence in calcium ion transport, demethylation, cell signaling, cell cycle and development. Continued sequencing of pCMS trios will identify additional risk genes and provide greater insights into biological mechanisms of stereotypies across diagnostic boundaries.

Citation: Fernandez TV, Williams ZP, Kline T, Rajendran S, Augustine F, Wright N, et al. (2023) Primary complex motor stereotypies are associated with de novo damaging DNA coding mutations that identify KDM5B as a risk gene. PLoS ONE 18(10): e0291978. https://doi.org/10.1371/journal.pone.0291978

Editor: Claudia Brogna, Fondazione Policlinico Universitario Gemelli IRCCS, ITALY

Received: April 13, 2023; Accepted: September 10, 2023; Published: October 3, 2023

Copyright: © 2023 Fernandez et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Data Availability: The data underlying the results presented in this study have been submitted to the Dryad Digital Repository and assigned a unique DOI (doi:10.5061/dryad.rfj6q57d5). The data submission is currently in “private for peer review” status, so this DOI will not be live until the manuscript is accepted for publication. However, a private URL to this data is provided by Dryad for use during peer review: https://datadryad.org/stash/share/HU8cwTlay7QNbwTWYhuuBiZcSeV75dtgmkzsB4B08N0 Clicking this link immediately launches a download of the data files in the repository.

Funding: This work was supported by grants from the Simons Foundation (sfari.org, SFARI award #239013, TVF), the Allison Family Foundation (TVF), Nesbitt-McMaster Foundation (HSS), Klump Family (HSS), and Graves Family (HSS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. There was no additional external funding received for this study.

Competing interests: I have read the journal’s policy, and the authors of this manuscript have the following competing interests: Dr. Fernandez receives research/grant support from the National Institutes of Mental Health. Dr. Olfson receives research support from the National Institutes of Mental Health, the Alan B. Slifka Foundation through the Riva Ariella Ritvo endowment, and the International Obsessive-Compulsive Disorder Foundation. Dr. Singer serves as a consultant for Abide Therapeutics, Inc; Cello Health BioConsulting; ClearView Healthcare Partners; Teva Pharmaceutical Industries Ltd; and Trinity Partners, LLC. Dr. Singer receives publishing royalties from Elsevier and research/grant support from the Tourette Association of America. Other authors declare no potential conflicts. This does not alter our adherence to PLOS ONE policies on sharing data and materials.

Introduction

Motor stereotypies are rhythmic, repetitive, prolonged, fixed, patterned, non-goal-directed movements that are often bilateral and temporarily stop with distraction. Complex motor stereotypies (CMS) include hand flapping, finger wiggling, head nodding, and rocking; these are often accompanied by mouth opening, head posturing, jumping, pacing, and occasional vocalizations [1]. Movements occur for up to minutes in duration, multiple times per day, and tend to be exacerbated by excitement, fatigue, stress, boredom, or being engrossed in an activity. CMS are common in children with autism spectrum disorder (ASD), intellectual disability, or sensory deprivation, as well as in typically developing children. A favored classification subdivides by etiology into primary (otherwise typically developing) and secondary categories. In both groups, stereotypies often result in social stigmatization, classroom disruption, and interference with academic activities.

In children with ASD, stereotypic behaviors (“secondary” stereotypies) occur in about 44% of patients and are recognized as a core phenotype of the disorder [2]. The severity and frequency of motor stereotypies is correlated with severity of illness, degree of intellectual disability, and impairments in adaptive functioning and symbolic play [3–9]. They are often associated with self-injurious behaviors [10, 11]. A wide range of medications have been tried for treatment of stereotypies in ASD, but efficacy is inconsistent and inadequate, with potential for long-term side effects [12].

Motor stereotypies also occur in otherwise typically developing children (“primary” stereotypies) [13–22]. Studies comparing primary and secondary stereotypies show that there is considerable similarity in their phenomenology [23–25]. Primary CMS (pCMS) has a typical age of onset before 3 years, and greater than 90% of children continue to experience CMS into adolescence and adulthood [16, 26, 27]. The prevalence of pCMS is estimated to be 3–4% of children in the U.S. [17, 26]. Similar to secondary stereotypies, medications are generally regarded as ineffective for primary CMS [13, 27], but there is evidence to support the benefits of cognitive behavioral therapy [28–30].

The precise pathophysiological mechanism for motor stereotypies remains obscure [31], though investigators have hypothesized abnormalities within cortico-striatal-thalamo-cortical pathways [32–38] and several neurotransmitter systems [33, 39–41]. A recent study reported reduced functional connectivity between prefrontal cortical and striatal regions in pCMS [42]. A genetic etiology for stereotypies has been suggested in primary and secondary categories, although the specific gene(s) contributing to this movement disorder remain unclear. With respect to secondary stereotypies in ASD, family studies have demonstrated that these repetitive behaviors are highly heritable, with a genetic etiology that is likely independent from other core diagnostic features [43]. While there are no studies of recurrence risk or twin concordance reported for pCMS, a positive family history is reported in 25–40%, while remaining cases appear to be sporadic [16, 27, 44].

Considering these findings, we conducted the first pilot genetic study of pCMS in 129 typically developing children and their parents. We hypothesized that pCMS may represent a more genetically homogenous group of individuals versus those with secondary stereotypies, thereby facilitating genetic discovery and insight into the biology of stereotypies more generally [48, 49]. We studied rare de novo, or spontaneous, germline DNA mutations in these individuals. In disorders such as autism, obsessive-compulsive disorder, and Tourette syndrome [45–47], this approach has proven invaluable for identifying genetic variants of large effect, high confidence risk genes, and enriched biological functions. Using whole-exome DNA sequencing, we identified an enrichment of de novo predicted-damaging coding mutations in pCMS and identified one high-confidence risk gene, Lysine Demethylase 5B (KDM5B) in our cohort. By further analysis of de novo damaging mutations in pCMS, we predict that there are approximately 184 pCMS risk genes and that sequencing more pCMS parent-child trios is a definite path toward discovering these genes. In this pilot study, we see a significant overlap between genes harboring de novo damaging mutations in pCMS and those in ASD as well as Tourette syndrome, a neurodevelopmental movement disorder characterized by motor and vocal tics. This overlap occurred despite excluding subjects with ASD or tics. Furthermore, owing to the two de novo damaging KDM5B mutations in our pCMS cohort, there is significant genetic overlap with ASD probands with highest stereotypy scores, but not those with low scores. Exploratory systems analyses of genes harboring de novo damaging mutations in pCMS show these genes to have peak expression in the cortex and striatum during early mid-fetal development. Finally, exploratory gene ontology analysis highlights functional convergence in calcium ion signaling, demethylation, cell signaling, cell cycle and development.

Materials and methods

Fig 1 provides an overview of the study methods.

thumbnail
Download:
Fig 1. Overview of variant discovery and data analysis.

We performed whole-exome DNA sequencing of 129 pCMS and 853 control parent-child trios. After quality control, 118 pCMS and 750 control trios remained for subsequent analyses. We performed a burden analysis, comparing the rates of de novo single nucleotide (SNVs) and insertion-deletion (indel) DNA variants between cases and controls. Next, we assessed the statistical significance of gene-level recurrence of de novo damaging variants in our pCMS group, identifying one high-confidence risk gene. Using the maximum likelihood estimation (MLE) method, de novo variant simulations, and TADA, we estimated the number of genes contributing to pCMS risk and used this estimate to predict the number of risk genes that will be discovered as more pCMS trios are sequenced. Finally, exploratory gene enrichment analyses were performed, assessing degree of overlap with gene sets harboring de novo damaging variants in other disorders, gene ontology terms, networks, and expression pattern clustering within certain brain regions across development.

https://doi.org/10.1371/journal.pone.0291978.g001

Subjects and assessment measures

This protocol was approved by the Johns Hopkins Medicine Institutional Review Board. Children with primary complex motor stereotypies (pCMS) were recruited from either the Johns Hopkins Pediatric Neurology Movement Disorder Outpatient Clinic (HSS, Director), or via email (singerlab@jhmi.edu). All participants verbally consented and provided signed parental consent. Using standardized forms via telephone, the study coordinator completed a brief screening general history, obtained baseline data about each child’s stereotypies, and completed an Autism Spectrum Screening Questionnaire (ASSQ). The presence of stereotypic movements was confirmed, either via direct observation in clinic or by video review (HSS). If the subject passed the screening assessment, additional data was collected on the child and both parents via RedCap, an electronic web-based application for data capture and online questionnaires. The latter included the Stereotypy Severity Scale (Motor and Impairment scores) and comorbidity measures (Multidimensional Anxiety Scale for Children—MASC; ADHD-Rating Scale IV; Conner’s Parent Rating Scale—CPRS; Repetitive Behavior Scale-Revised—RBS-R; Children’s Yale-Brown Obsessive-Compulsive Scale—CYBOCS; and Social Responsiveness Scale—SRS) (see S1 File).

For this pilot study, we prioritized the study of “simplex” pCMS (children without known family history of affected first or second-degree relatives) to increase the likelihood of detecting de novo DNA sequence variants. Eligibility required participants to have: (a) confirmed complex motor stereotypies; (b) onset before age 3 years; (c) temporary suspension of movements by an external stimulus or distraction. Exclusion criteria included: (a) a total score >13 on the ASSQ or a prior autism spectrum disorder diagnosis; (b) historical evidence supporting the absence of intellectual disability; (c) seizures or a known neurological disorder; and (d) the presence of motor/vocal tics. The presence of inattentiveness, hyperactivity, or impulsivity (i.e., ADHD symptoms) and/or obsessive-compulsive behaviors were not exclusionary.

DNA whole-exome sequencing (WES)

DNA was collected from all children meeting eligibility criteria and from their parents, using the Oragene OG-500 collection kit and standard extraction protocols (DNA Genotek, Ottowa, Ontario, Canada). Exome capture and sequencing were performed at the Yale Center for Genome Analysis (YCGA), using the NimbleGen SeqCap EZExomeV2 capture library (Roche NimbleGen, Madison, WI, USA) and the Illumina HiSeq 2500 platform (Illumina, San Diego, CA, USA). WES data from 853 unaffected parent-child trios (2,559 samples total) were obtained from the Simons Simplex Collection via the NIH Data Archive (https://ndar.nih.gov/edit_collection.html?id=2042). These children and their parents have no evidence of autism spectrum or other neurodevelopmental disorders [48]. The same exome capture and sequencing platforms were used for these control samples.

Sequence alignment, variant calling, and quality control

Alignment and variant calling of the sequencing reads followed the latest Genome Analysis Toolkit (GATK) [49] Best Practices guidelines, as described previously [46]. Variants were annotated using RefSeq hg19 gene definitions using ANNOVAR [50]. Trios were omitted from downstream analyses if (a) genetic markers were not consistent with expected family relationships; (b) an excessive number of de novo variants were observed, or (c) if they were outliers in principal components analysis (see S1 File). De novo variants were called and confirmed as previously described [46] and as detailed in S1 File.

Mutation rate and gene recurrence

Within each cohort, we calculated the rate of de novo DNA mutations per base pair, using methods previously described [46]. We included only those de novo variants present with a frequency of <0.001 (0.1%) in the ExAC v0.3.1 database [51] and compared de novo mutation rates in cases versus controls using a one-tailed rate ratio test (S1 File). Because our cases and controls were sequenced at different times, we took precautions to ensure that batch effects, including differences in sequencing depth and quality, did not influence our comparisons. First, we compared cases and controls that were sequenced on the same sequencing platform and using the same capture library. Second, we considered only “callable” bases, defined as loci with ≥ 20x sequencing depth in all family members, with base quality ≥ 20, and map quality ≥ 30; these thresholds match those required for GATK and de novo variant calling. Third, for each cohort, we summed the “callable” base pairs in every family and used this number as the denominator for de novo rate calculations. In this way, we normalized the de novo rates to guard against any residual differences in sequencing depth or quality, and we compared these normalized rates between cases and controls. This method of comparing different batches of sequencing data has been used in several prior studies [45, 46, 52, 53].

As described in our previous WES studies [45, 46, 52], we used the Transmitted And De novo Association (TADA-Denovo) test as a statistical method for risk gene discovery based on gene-level recurrence of de novo mutations within the classes of variants that we found enriched in pCMS [54, 55]. This test generates random mutational data based on each gene’s specified mutation rate to determine null distributions, then calculates a p-value and a false discovery rate (FDR) q-value for each gene using a Bayesian "direct posterior approach." A low q-value represents strong evidence for pCMS association. See S1 File for details.

Estimating the number of pCMS risk genes

As described previously [46, 52], we used a maximum likelihood estimation (MLE) method [56] to estimate the number of genes contributing risk to pCMS, based on the observed number of de novo damaging variants in our dataset. See S1 File for details of these calculations.

Next, we used previously described methods [46, 52] to predict the likely number of risk genes that will be discovered as additional pCMS parent-child trios are sequenced by WES. These predictions utilize the estimated number of pCMS risk genes along with pCMS de novo mutation rates observed in our study to perform mutation simulations, followed by TADA-Denovo testing (see S1 File).

Gene set overlap

We used DNENRICH [57] (https://statgen.bitbucket.io/dnenrich/index.html) to test whether genes harboring de novo damaging mutations in our pCMS subjects were significantly enriched among genes harboring de novo damaging mutations in several neuropsychiatric disorders, including autism (ASD), schizophrenia (SCZ), Tourette’s disorder (TD), obsessive-compulsive disorder (OCD), developmental disorders (DD), intellectual disability (ID), and epileptic encephalopathy (EE). Additionally, we were interested in the question of whether our pCMS cohort share genes harboring de novo damaging mutations with ASD probands having high versus low stereotypy scores. To approach this question, we assembled lists of genes harboring de novo damaging mutations in ASD probands from the Simons Simplex Collection (SSC) for whom stereotyped behavior scores (Stereotyped Behavior Score from the RBS-R, Repetitive Behavior Scale-Revised) were available. We looked for overlap between our pCMS cohort and those SSC ASD probands with stereotypy scores in the 90th percentile (high stereotypies) and those scoring in the 10th percentile (low stereotypies). These gene lists are compiled in S4 Table. Further details about gene list curation and DNENRICH methods can be found in S1 File.

Exploratory gene ontology, network, and spatiotemporal analyses

To determine whether genes harboring de novo damaging variants in pCMS may perform similar biological functions, we used the list of pCMS genes harboring de novo damaging mutations to identify overlap with gene ontologies using two tools: Enrichr (https://maayanlab.cloud/Enrichr/) [58] and ConsensusPathDB (http://cpdb.molgen.mpg.de/). We identified gene ontology and pathway terms with an enrichment p-value < 0.05. We also used Ingenuity Pathway Analysis (IPA, Ingenuity Systems, http://www.ingenuity.com/) to identify potential gene networks based on this same gene list with the lowest likelihood of interactions due to chance.

Finally, using this same list of genes harboring de novo damaging variants in pCMS, we searched for possible enrichment of gene expression within certain brain regions across multiple developmental time periods, using data from the Brainspan Atlas of the Developing Human Brain [59, 60]. See S1 File.

Results

We performed WES on 129 pCMS parent-child trios (387 samples total) meeting inclusion criteria. WES data from 853 unaffected control trios, already sequenced from the Simons Simplex Collection, were pooled with our pCMS trios for joint variant calling. After quality control methods, our sample size for a burden analysis was 118 pCMS and 750 unaffected trios (Table 1, Fig 1, S1 Table, S1 Fig).

thumbnail
Download:
Table 1. Distribution of de novo variants in pCMS cases and controls.

https://doi.org/10.1371/journal.pone.0291978.t001

Increased burden of de novo damaging variants in pCMS

Based on work in other neurodevelopmental disorders, we expected to find an enrichment of de novo likely gene disrupting (LGD) variants (stop codon, frameshift, or canonical splice-site variants) in pCMS probands versus controls. We found a statistically significant increased rate of de novo LGD variants in pCMS cases, confirming our hypothesis (rate ratio [RR] 1.95, 95% Confidence Interval [CI] 1.04–3.50, p = 0.04). Furthermore, de novo variants predicted to be damaging (LGD plus missense variants with Polyphen2-HDIV score <0.957 and ≥0.453) were also over-represented in pCMS probands (RR 1.37, CI 1.05–1.76, p = 0.03). We did not detect a difference in mutation rates for de novo synonymous variants, or when all de novo variants (coding +/- non-coding) were considered together (Table 1, Fig 2, S2 Table).

thumbnail
Download:
Fig 2. Rates of de novo variants in pCMS cases versus controls.

Bar chart comparing the rates of de novo variant classes between pCMS cases (red) and controls (blue). Comparisons are between per base pair (bp) mutation rates, using a one-tailed rate ratio test. Statistically significant comparisons (p<0.05) are marked with asterisks. Error bars show 95% confidence intervals.

https://doi.org/10.1371/journal.pone.0291978.g002

KDM5B is a high-confidence candidate risk gene in pCMS

Having established a higher rate of de novo damaging variants in pCMS probands, we next asked whether these variants cluster within specific genes. We identified one gene with more than one predicted damaging de novo variant in unrelated probands: KDM5B (Lysine Demethylase 5B) harbored two different LGD (stopgain) de novo variants in pCMS probands 1029–03 and 1050–03. Using TADA-Denovo [54] and previously established false discovery rate (FDR) thresholds, we found that KDM5B meets statistical criteria for a high-confidence risk gene (q<0.1) in pCMS (S3 Table).

Approximately 184 genes contribute to pCMS risk

Based on the number of observed de novo damaging mutations in pCMS, the MLE method estimated the most likely number of pCMS risk genes to be 184 (S2 Fig). Next, we used this estimate along with de novo mutation rates observed in pCMS trios to predict the likely number of these 184 risk genes that will be discovered in larger pCMS cohorts. Based on these simulations, WES of 500 trios should find 16 probable and 7 high-confidence risk genes; 1000 trios should find 51 probable and 26 high-confidence risk genes (S3 Fig).

pCMS gene enrichment in Tourette’s disorder, ASD, and in ASD with high versus low stereotypy scores

Using DNENRICH [57], we found significant overlap between genes harboring de novo damaging variants in pCMS (52 genes after excluding two genes with de novo damaging variants in controls, Table 2, S2 Table) and several gene sets curated from the literature (S4 Table). In particular, our pCMS cohort genes show significant gene overlap with autism probands with high stereotypy scores (5.8x enrichment, p = 0.047), Tourette’s disorder (4.5x enrichment, p = 0.019), autism spectrum disorder (2.2x enrichment, p = 0.0055–0.0069). There was no significant overlap with OCD, schizophrenia, intellectual disability, developmental disorders, or epileptic encephalopathy (S4 Table).

thumbnail
Download:
Table 2. 52 Genes harboring de novo damaging variants in pCMS and not in controls.

https://doi.org/10.1371/journal.pone.0291978.t002

Exploratory gene ontology, network, and spatiotemporal analyses

Using this same list of 52 genes harboring de novo damaging variants in pCMS (Table 2, S2 Table), we performed exploratory analyses to identify enrichment in biological, cellular, and molecular gene ontology terms. Using two enrichment tools, we identified significant enrichment for calcium ion transport and demethylation (adjusted p-value < 0.05 in either tool). By relaxing the statistical threshold to an unadjusted p < 0.05, we identified enrichment for these same gene ontology terms in results from both tools (S5 Table). Finally, we performed an exploratory gene network analysis of these 52 genes using IPA and identified the potential importance of these genes in cell signaling, cellular assembly and organization (S5 Table). Finally, mapping our pCMS de novo damaging variant genes onto the Brainspan Atlas of the Developing Human Brain gene expression data, we see nominal enrichment of gene expression in early mid-fetal cortex and striatum, with a trend toward enrichment in early fetal hippocampus, late mid-fetal cerebellum, and young childhood cerebellum (S5 Table).

Discussion

Like prior studies of ASD, Tourette’s disorder, and OCD, the current study demonstrates that the identification of de novo DNA coding variants will identify risk genes and provide a reliable entry-point into understanding the biology of stereotypies. We are studying otherwise typically developing children with stereotypies (primary CMS), as this may represent a more genetically homogenous group of individuals versus those with secondary stereotypies, thereby facilitating genetic discovery and insight into the biology of stereotypies more generally [61, 62]. Despite our small cohort size, we identified two de novo nonsense mutations in KDM5B in unrelated probands, and we show that finding two such independent mutations in our cohort is highly unlikely to be a chance occurrence.

KDM5B is a lysine-specific demethylase that removes methyl groups from tri-, di- and monomethylated lysine 4 on histone 3. KDM5B acts a transcriptional repressor and has primarily been implicated in the pathogenesis of cancer [63]. More recently, this gene has also been implicated in congenital heart disease risk, embryonic development, DNA repair, adult cognitive function, and muscle strength [64–69]. KDM5B has been identified as high-confidence risk gene in ASD via detection of heterozygous de novo damaging variants in WES studies [47, 55] and for developmental disorders more broadly [70]. Individuals reported in the literature with intellectual disability/developmental delay harboring KDM5B mutations often show an autosomal recessive inheritance pattern, including inherited homozygous or compound heterozygous mutations in this gene [71, 72], while heterozygous mutations occur more frequently in probands from the Deciphering Developmental Disorders Study [73]. A recent study by Chen et al. [68] identified KDM5B as one of eight genes associated with adult cognitive function through rare protein-truncating and damaging missense variants. Consistent with prior reports, they identified a gene dose effect, whereby individuals with rare heterozygous protein-truncating variants showed higher adult cognitive function measurements compared to those with homozygous damaging mutations [68]. Both this study and another recent report by Huang et al. [69] found a significant association between rare variants in KDM5B and hand grip strength, a phenotype related to muscle function, and one of several additional phenotypes found to be associated with KDM5B [68]. Interestingly, KDM5B has a relatively high rate of protein truncating variants, with a rate of approximately 1 in 1,900 subjects in the UK Biobank sample. This is in contrast to most other genes linked to neurodevelopmental phenotypes. Considering the substantial pleiotropic and gene dosage effects reported in several studies to date, it is interesting to see how different mutations and inheritance patterns in this gene can lead to a spectrum of phenotypic outcomes, including ASD, ID/DD, congenital heart disease, adult cognitive function, muscle strength, and now pCMS in childhood. Our team subsequently interviewed families harboring KDM5B mutations in our pCMS child probands and confirmed that there was no evidence for ASD, ID, or congenital heart disease.

Expression of KDM5B is normally restricted to the brain and the testis [74]. Within the brain, high expression levels of KDM5B are seen in the cerebellum (S4 Fig), and expression across all brain regions is highest prenatally (S5 Fig). Consistent with this data, a recent MRI study from our group found volumetric differences in the cerebellum of children with pCMS versus controls, and these changes correlated with Stereotypy Severity Scores [75]. Similarly, cerebellar volume was correlated with stereotyped activity in a deer mouse animal model with repetitive behaviors [75]. The identification of this risk gene in pCMS suggests that chromatin (dys)regulation of KDM5B target genes may be one contributing mechanism underlying stereotypies across diagnostic boundaries. Further studies are warranted to determine the downstream effects of these mutations in the developing brain. These studies are underway in our laboratory.

It is interesting that we find significant overlap between genes harboring de novo damaging mutations in pCMS and those reported in a recent study of Tourette syndrome (S4 Table; 4.5x enrichment, p = 0.019). While we have reported approximately 25% of pCMS patients have co-existing tics [14], we find this overlap with Tourette despite excluding pCMS subjects with co-existing motor or vocal tics (see Methods). Enriched expression of pCMS genes in the cortex and striatum (S5 Table) is also consistent with widely believed involvement of these regions in Tourette syndrome. While OCD was not exclusionary in our pCMS study, we saw no significant gene overlap with OCD. Similarly, we found no significant overlap with SCZ, ID, DD, or EE. We did, however, find significant overlap between pCMS and ASD risk genes (2.2x enrichment, p = 0.006–0.007), despite no evidence of ASD in our subjects.

With regard to stereotypies in ASD, we curated lists of genes harboring de novo damaging mutations in SSC probands with the highest (90th percentile) and lowest (10th percentile) stereotypies, measured by Stereotyped Behavior Scores (SBS) from the RBS-R. KDM5B mutations were found only in SSC probands with high stereotypy scores, yielding 5.8-fold enrichment over expectation (p = 0.047) when compared against our pCMS genes (S4 Table). To further examine the relation between de novo KDM5B mutations stereotypies in SSC ASD probands, we compared SBS scores in four probands with KDM5B mutations versus 364 age-matched patients without (S6 Fig). Scores were higher in mutation carriers, but this did not reach statistical significance (p = 0.076), likely due to the low number of mutation carriers in this cohort.

In summary, we report an increased burden of de novo damaging heterozygous DNA coding variants in primary complex motor stereotypies. We identified one high-confidence risk gene for pCMS in our pilot cohort and estimate that there are 184 genes conferring risk for this phenotype. Whole-exome sequencing in parent-child pCMS trios provides a reliable way to make progress in gene discovery. Our exploratory analyses of genes harboring de novo damaging mutations in pCMS highlight several gene ontology terms (comprising biological processes, molecular functions, and cellular components), as well as brain regions and developmental time periods. These preliminary findings provide insights into possible etiologies of stereotypies, and this knowledge is a prerequisite for developing new treatments. Further sequencing and mechanistic studies are warranted to understand this phenotype, which has relevance across diagnostic boundaries.

Supporting information

S1 Fig. PCA scree and individual plots.

Scree plots following Principal Components Analysis (PCA), showing (A) the percentage of variance captured by each of the first 32 principal components, and (B) the cumulative percentage of variance captured by these same components in the exome metrics data from cases and controls. The “elbow” of the scree plot is visualized to be around the 5th principal component. This was confirmed by the Factominer R code function “estim_ncp()”. The first 5 PCs capture over 80% of the variance, and this number of PCs was used to determine PCA outliers during quality control (see S1 Table and S1 File). (C) Individual plots for the first two principal components, based on PCA of exome sequencing quality metrics. pCMS cases are plotted in red, and controls in blue. The first two PCs together capture 56.3% of the variance. R code to generate this data and figure are in S1 File, and individual PC factor values are in S1 Table. This figure includes PCA outliers (>3 standard deviations from the mean in PCs 1–5), which were removed during quality control, prior to further analysis of case-control data.

https://doi.org/10.1371/journal.pone.0291978.s001

(TIF)

S2 Fig. Maximum Likelihood Estimate (MLE) of number of pCMS risk genes.

For each number of possible risk genes between 1–2,500, we conducted 50,000 simulations to determine the number of risk genes that yielded the closest agreement between our observed and simulated data. This MLE method yields an estimate of 184 pCMS risk genes (red vertical line). See S1 File.

https://doi.org/10.1371/journal.pone.0291978.s002

(TIF)

S3 Fig. Gene discovery by number of trios sequenced.

Using our estimate of 184 risk genes (based on the MLE method–see Main Text and S1 File), we estimated the number of probable (FDR q<0.3) and high-confidence (FDR q<0.1) risk genes that will be discovered as more pCMS trios are sequenced. We performed 10,000 simulations at each cohort size from 25–3,000 trios, randomly generating variants and assigning to risk genes in agreement with the proportions seen in our data, then applying the TADA-Denovo algorithm. Based on these simulations, WES of 500 trios should find 16 probable and 7 high-confidence risk genes; 1000 trios should find 51 probable and 26 high-confidence risk genes.

https://doi.org/10.1371/journal.pone.0291978.s003

(TIF)

S4 Fig. KDM5B brain expression levels.

Brain expression by region for KDM5B. Data is from GTEx Analysis Release V7 (dbGaP Accession phs000424.v7.p2) (https://gtexportal.org/home/gene/KDM5B). Expression values are shown in Transcripts Per Million (TPM), calculated from a gene model with isoforms collapsed to a single gene. No other normalization steps have been applied. Box plots are shown as median, 25th, and 75th percentiles. Points are displayed as outliers if they are above or below 1.5 times the interquartile range. Further details about expression quantification and samples can be found at https://gtexportal.org/home/documentationPage#AboutData.

https://doi.org/10.1371/journal.pone.0291978.s004

(TIF)

S5 Fig. KDM5B spatiotemporal brain expression.

Brain expression trajectories of KDM5B in the developing human brain. Expression data is from the Brainspan Consortium (brainspan.org, hbatlas.org), generated using the Affymetrix GeneChip Human Exon 1.0 ST Array platform. Vertical axis is the log2-transformed array signal intensity, which is proportional to transcript expression. A stringent threshold of ≥6 was required to meet criteria for brain expression. Horizontal axis represents periods of human development and adulthood as previously defined by Kang et al (2011). Birth begins period 8 and adolescence begins period 12. Brain regions are by color: neocortex (NCX), hippocampus (HIP), amygdala (AMY), striatum (STR), mediodorsal nucleus of the thalamus (MD), cerebellar cortex (CBC).

https://doi.org/10.1371/journal.pone.0291978.s005

(TIF)

S6 Fig. Stereotypy scores in Simons Simplex Collection ASD probands.

Tukey box and whisker plot of Stereotyped Behavior Score (SBS) from the RBS-R (Repetitive Behavior Scale-Revised) in aged-matched Simons Simplex Collection ASD probands with (+, n = 4) and without (-, n = 364) de novo damaging mutations in KDM5B. Two-tailed Mann-Whitney test of ages (months) between groups: p = 0.86. One-tailed Mann-Whitney test of SBS between groups: p = 0.076.

https://doi.org/10.1371/journal.pone.0291978.s006

(TIF)

S1 Table. Phenotype, exome sequencing metrics, and principal components analysis.

See “S1 Table”. First tab contains individual-level sample information (columns A-I), including family ID, individual ID, phenotype, cohort, collection site, gender, capture platform, size of “callable exome”, and paternal age (years) at birth, where available. Column J lists reasons for any sample exclusions by quality control methods; “0” indicates that the sample was not excluded and was included in subsequent analyses. Columns K-AF list individual sample sequencing metrics generated using PicardTools, and GATK DepthOfCoverage tools. Columns AG-AQ list individual sample sequencing metrics generated using PLINK/SEQ (i-stats; https://psychgen.u.hpc.mssm.edu/plinkseq/stats.shtml). Columns B, K-AQ were included in Principal Components Analysis (PCA). Third tab contains cohort-level metrics calculated using samples passing quality control. ±95% confidence intervals are given, when applicable. Fourth tab contains coordinates generated for each sample for the top 10 principal components following PCA. The code used to generate this data is included in S1 File. Using these coordinates, we removed trios with family members falling more than three standard deviations from the mean in any of the first five principal components; this information is contained in the fifth tab.

https://doi.org/10.1371/journal.pone.0291978.s007

(XLSX)

S2 Table. Annotated de novo variants in pCMS and controls.

See “S2 Table”. Detailed information on all high confidence de novo variants in cases and controls. These variants were annotated using ANNOVAR, based on RefSeq hg19 gene definitions. Column descriptions are provided in a separate tab of this file.

https://doi.org/10.1371/journal.pone.0291978.s008

(XLSX)

S3 Table. Gene-level de novo mutation rates, variant counts, and TADA-Denovo results.

See “S3 Table”. First tab contains de novo mutation rates used to perform subsequent maximum likelihood estimation (MLE) and TADA-Denovo analyses. The following mutation rates are listed for each gene: overall (mut.rate), likely gene disrupting (lgd), predicted damaging missense (mis3, also referred to as Mis-D), and all damaging (lgd + mis3). These overall mutation rates were previously published (Ware et al., 2015) from unaffected parent-child trios. The code used to generate the mutation rate table is provided in S1 File. Second tab contains the input file for the TADA-Denovo algorithm. Gene-level expected mutation rates for LGD (“mut.cls1” column) and Mis-D variants (“mut.cls2” column) are listed, along with their respective observed mutation counts in our pCMS data (“dn.cls1” and “dn.cls2”, respectively). Code for running TADA-Denovo is given in S1 File. Third tab contains the final output results from TADA-Denovo code provided in S1 File. One gene harboring more than one damaging de novo (LGD or Mis-D) variant in unrelated pCMS families is highlighted in yellow (KDM5B). This gene exceeded the threshold for being considered a high confidence (qval < 0.1) risk gene.

https://doi.org/10.1371/journal.pone.0291978.s009

(XLSX)

S4 Table. DNENRICH gene lists and results.

See “S4 Table”. See S1 File for details of DNENRICH analysis and gene lists used. First tab contains input gene lists and information about their curation. Second tab contains the input mutation list for DNENRICH; each row represents a de novo damaging mutation in a pCMS proband. Third tab contains results output from DNENRICH. Significantly enriched gene sets are highlighted.

https://doi.org/10.1371/journal.pone.0291978.s010

(XLSX)

S5 Table. Exploratory pathway, gene ontology, and spatiotemporal analyses results.

See “S5 Table”. Gene ontology results from Enrichr are in the first tab, including unadjusted and adjusted p-values. Gene ontology results from ConsensusPathDB are in the second tab; p-values < 0.05 and corresponding q-values are shown. Network analysis results from IPA are shown in the third tab. Specific Enrichment Analysis (SEA) exploring whether genes cluster within certain brain regions across development using Brainspan atlas data is in the fourth tab; p-values < 0.05 are highlighted in yellow. See S1 File for details of these analyses.

https://doi.org/10.1371/journal.pone.0291978.s011

(XLSX)

S1 File. Supplementary methods.

https://doi.org/10.1371/journal.pone.0291978.s012

(DOCX)

Acknowledgments

We wish to thank the families who have participated in and contributed to this study. Control subject data were obtained from the NIH-supported National Database for Autism Research (NDAR). NDAR is a collaborative informatics system created by the National Institutes of Health to provide a national resource to support and accelerate research in autism. Dataset identifier: 2042. This manuscript reflects the views of the authors and may not reflect the opinions or views of the NIH or of the Submitters submitting original data to NDAR. We are grateful to all of the families at the participating Simons Simplex Collection (SSC) sites, as well as the principal investigators (A. Beaudet, R. Bernier, J. Constantino, E. Cook, E. Fombonne, D. Geschwind, R. Goin-Kochel, E. Hanson, D. Grice, A. Klin, D. Ledbetter, C. Lord, C. Martin, D. Martin, R. Maxim, J. Miles, O. Ousley, K. Pelphrey, B. Peterson, J. Piggot, C. Saulnier, M. State, W. Stone, J. Sutcliffe, C. Walsh, Z. Warren, E. Wijsman). We appreciate obtaining access to phenotype and genetic data on SFARI Base. Approved researchers can obtain the SSC population dataset described in this study (https://www.sfari.org/resource/simons-simplex-collection/) by applying at https://base.sfari.org.

References

  1. 1. Singer HS. Stereotypic movement disorders. Handbook of clinical neurology. 2011;100:631–9. pmid:21496612
  2. 2. American Psychiatric Association. Diagnostic and statistical manual of mental disorders: DSM-5. 2013.
  3. 3. Militerni R, Bravaccio C, Falco C, Fico C, Palermo MT. Repetitive behaviors in autistic disorder. Eur Child Adolesc Psychiatry. 2002;11(5):210–8. pmid:12469238
  4. 4. Bishop SL, Richler J, Lord C. Association between restricted and repetitive behaviors and nonverbal IQ in children with autism spectrum disorders. Child Neuropsychol. 2006;12(4–5):247–67. pmid:16911971
  5. 5. Goldman S, Wang C, Salgado MW, Greene PE, Kim M, Rapin I. Motor stereotypies in children with autism and other developmental disorders. Dev Med Child Neurol. 2009;51(1):30–8. pmid:19087102
  6. 6. Honey E, Leekam S, Turner M, McConachie H. Repetitive behaviour and play in typically developing children and children with autism spectrum disorders. J Autism Dev Disord. 2007;37(6):1107–15. pmid:17072748
  7. 7. Campbell M, Locascio JJ, Choroco MC, Spencer EK, Malone RP, Kafantaris V, et al. Stereotypies and tardive dyskinesia: abnormal movements in autistic children. Psychopharmacol Bull. 1990;26(2):260–6. pmid:2236468
  8. 8. Bodfish JW, Symons FJ, Parker DE, Lewis MH. Varieties of repetitive behavior in autism: comparisons to mental retardation. J Autism Dev Disord. 2000;30(3):237–43. pmid:11055459
  9. 9. Matson JL, Kiely SL, Bamburg JW. The effect of stereotypies on adaptive skills as assessed with the DASH-II and Vineland Adaptive Behavior Scales. Res Dev Disabil. 1997;18(6):471–6. pmid:9403929
  10. 10. Gal E, Dyck MJ, Passmore A. The relationship between stereotyped movements and self-injurious behavior in children with developmental or sensory disabilities. Res Dev Disabil. 2009;30(2):342–52. pmid:18693081
  11. 11. Symons FJ, Sperry LA, Dropik PL, Bodfish JW. The early development of stereotypy and self-injury: a review of research methods. Journal of intellectual disability research: JIDR. 2005;49(Pt 2):144–58. pmid:15634323
  12. 12. Doyle CA, McDougle CJ. Pharmacologic treatments for the behavioral symptoms associated with autism spectrum disorders across the lifespan. Dialogues in clinical neuroscience. 2012;14(3):263–79. pmid:23226952
  13. 13. Tan A, Salgado M, Fahn S. The characterization and outcome of stereotypical movements in nonautistic children. Mov Disord. 1997;12(1):47–52. pmid:8990053
  14. 14. Mahone EM, Bridges D, Prahme C, Singer HS. Repetitive arm and hand movements (complex motor stereotypies) in children. The Journal of pediatrics. 2004;145(3):391–5. pmid:15343197
  15. 15. Leekam S, Tandos J, McConachie H, Meins E, Parkinson K, Wright C, et al. Repetitive behaviours in typically developing 2-year-olds. J Child Psychol Psychiatry. 2007;48(11):1131–8. pmid:17995489
  16. 16. Harris KM, Mahone EM, Singer HS. Nonautistic motor stereotypies: clinical features and longitudinal follow-up. Pediatr Neurol. 2008;38(4):267–72. pmid:18358406
  17. 17. Foster LG. Nervous habits and stereotyped behaviors in preschool children. J Am Acad Child Adolesc Psychiatry. 1998;37(7):711–7. pmid:9666626
  18. 18. Rafaeli-Mor N, Foster L, Berkson G. Self-reported body-rocking and other habits in college students. Am J Ment Retard. 1999;104(1):1–10. pmid:9972830
  19. 19. MacDonald R, Green G, Mansfield R, Geckeler A, Gardenier N, Anderson J, et al. Stereotypy in young children with autism and typically developing children. Res Dev Disabil. 2007;28(3):266–77. pmid:16814515
  20. 20. Sallustro F, Atwell CW. Body rocking, head banging, and head rolling in normal children. The Journal of pediatrics. 1978;93(4):704–8. pmid:309000
  21. 21. Prevalence Tröster H. and functions of stereotyped behaviors in nonhandicapped children in residential care. J Abnorm Child Psychol. 1994;22(1):79–97.
  22. 22. Castellanos FX, Ritchie GF, Marsh WL, Rapoport JL. DSM-IV stereotypic movement disorder: persistence of stereotypies of infancy in intellectually normal adolescents and adults. J Clin Psychiatry. 1996;57(3):116–22. pmid:8617696
  23. 23. Baranek GT. Autism during infancy: a retrospective video analysis of sensory-motor and social behaviors at 9–12 months of age. J Autism Dev Disord. 1999;29(3):213–24. pmid:10425584
  24. 24. Werner E, Dawson G. Validation of the phenomenon of autistic regression using home videotapes. Arch Gen Psychiatry. 2005;62(8):889–95. pmid:16061766
  25. 25. Loh A, Soman T, Brian J, Bryson SE, Roberts W, Szatmari P, et al. Stereotyped motor behaviors associated with autism in high-risk infants: a pilot videotape analysis of a sibling sample. J Autism Dev Disord. 2007;37(1):25–36. pmid:17219059
  26. 26. Singer HS. Motor stereotypies. Seminars in pediatric neurology. 2009;16(2):77–81. pmid:19501335
  27. 27. Oakley C, Mahone EM, Morris-Berry C, Kline T, Singer HS. Primary complex motor stereotypies in older children and adolescents: clinical features and longitudinal follow-up. Pediatr Neurol. 2015;52(4):398–403.e1. pmid:25661287
  28. 28. Miller JM, Singer HS, Bridges DD, Waranch HR. Behavioral therapy for treatment of stereotypic movements in nonautistic children. Journal of child neurology. 2006;21(2):119–25. pmid:16566875
  29. 29. Specht MW, Mahone EM, Kline T, Waranch R, Brabson L, Thompson CB, et al. Efficacy of parent-delivered behavioral therapy for primary complex motor stereotypies. Dev Med Child Neurol. 2017;59(2):168–73. pmid:27259464
  30. 30. Singer HS, Rajendran S, Waranch HR, Mahone EM. Home-Based, Therapist-Assisted, Therapy for Young Children With Primary Complex Motor Stereotypies. Pediatr Neurol. 2018;85:51–7. pmid:30049425
  31. 31. Gao S, Singer HS. Complex motor stereotypies: an evolving neurobiological concept. Future Neurology. 2013;8(3):273–85.
  32. 32. Carter JC, Capone GT, Kaufmann WE. Neuroanatomic correlates of autism and stereotypy in children with Down syndrome. Neuroreport. 2008;19(6):653–6. pmid:18382280
  33. 33. Kelley AE, Lang CG, Gauthier AM. Induction of oral stereotypy following amphetamine microinjection into a discrete subregion of the striatum. Psychopharmacology (Berl). 1988;95(4):556–9. pmid:3145527
  34. 34. Kates WR, Lanham DC, Singer HS. Frontal white matter reductions in healthy males with complex stereotypies. Pediatr Neurol. 2005;32(2):109–12. pmid:15664770
  35. 35. Sato S, Hashimoto T, Nakamura A, Ikeda S. Stereotyped stepping associated with lesions in the bilateral medial frontoparietal cortices. Neurology. 2001;57(4):711–3. pmid:11524487
  36. 36. Maraganore DM, Lees AJ, Marsden CD. Complex stereotypies after right putaminal infarction: a case report. Mov Disord. 1991;6(4):358–61. pmid:1758457
  37. 37. Singer HS. Motor control, habits, complex motor stereotypies, and Tourette syndrome. Annals of the New York Academy of Sciences. 2013;1304:22–31. pmid:24175737
  38. 38. Mahone EM, Crocetti D, Tochen L, Kline T, Mostofsky SH, Singer HS. Anomalous Putamen Volume in Children With Complex Motor Stereotypies. Pediatr Neurol. 2016;65:59–63. pmid:27751663
  39. 39. Evans AH, Katzenschlager R, Paviour D, O’Sullivan JD, Appel S, Lawrence AD, et al. Punding in Parkinson’s disease: its relation to the dopamine dysregulation syndrome. Mov Disord. 2004;19(4):397–405. pmid:15077237
  40. 40. Lewis MH, Bodfish JW, Powell SB, Wiest K, Darling M, Golden RN. Plasma HVA in adults with mental retardation and stereotyped behavior: biochemical evidence for a dopamine deficiency model. Am J Ment Retard. 1996;100(4):413–8. pmid:8718995
  41. 41. Harris AD, Singer HS, Horska A, Kline T, Ryan M, Edden RA, et al. GABA and Glutamate in Children with Primary Complex Motor Stereotypies: An 1H-MRS Study at 7T. AJNR American journal of neuroradiology. 2016;37(3):552–7. pmid:26542237
  42. 42. Augustine F, Nebel MB, Mostofsky SH, Mahone EM, Singer HS. Aberrant prefrontal cortical-striatal functional connectivity in children with primary complex motor stereotypies. Cortex. 2021;142:272–82. pmid:34303880
  43. 43. Dworzynski K, Happé F, Bolton P, Ronald A. Relationship between symptom domains in autism spectrum disorders: a population based twin study. J Autism Dev Disord. 2009;39(8):1197–210. pmid:19373549
  44. 44. Freeman RD, Soltanifar A, Baer S. Stereotypic movement disorder: easily missed. Dev Med Child Neurol. 2010;52(8):733–8. pmid:20187883
  45. 45. Wang S, Mandell JD, Kumar Y, Sun N, Morris MT, Arbelaez J, et al. De Novo Sequence and Copy Number Variants Are Strongly Associated with Tourette Disorder and Implicate Cell Polarity in Pathogenesis. Cell reports. 2018;24(13):3441–54.e12. pmid:30257206
  46. 46. Cappi C, Oliphant ME, Peter Z, Zai G, Conceicao do Rosario M, Sullivan CAW, et al. De Novo Damaging DNA Coding Mutations Are Associated With Obsessive-Compulsive Disorder and Overlap With Tourette’s Disorder and Autism. Biol Psychiatry. 2020;87(12):1035–44. pmid:31771860
  47. 47. Satterstrom FK, Kosmicki JA, Wang J, Breen MS, De Rubeis S, An JY, et al. Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism. Cell. 2020;180(3):568–84 e23. pmid:31981491
  48. 48. Fischbach GD, Lord C. The Simons Simplex Collection: a resource for identification of autism genetic risk factors. Neuron. 2010;68(2):192–5. pmid:20955926
  49. 49. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, et al. The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data. Genome research. 2010;20(9):1297–303. pmid:20644199
  50. 50. Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic variants from high-throughput sequencing data. Nucleic acids research. 2010;38(16):e164. pmid:20601685
  51. 51. Lek M, Karczewski KJ, Minikel EV, Samocha KE, Banks E, Fennell T, et al. Analysis of protein-coding genetic variation in 60,706 humans. Nature. 2016;536(7616):285–91. pmid:27535533
  52. 52. Willsey AJ, Fernandez TV, Yu D, King RA, Dietrich A, Xing J, et al. De Novo Coding Variants Are Strongly Associated with Tourette Disorder. Neuron. 2017;94(3):486–99 e9. pmid:28472652
  53. 53. Olfson E, Lebowitz ER, Hommel G, Pashankar N, Silverman WK, Fernandez TV. Whole-exome DNA sequencing in childhood anxiety disorders identifies rare de novo damaging coding variants. Depress Anxiety. 2022;39(6):474–84. pmid:35312124
  54. 54. He X, Sanders SJ, Liu L, De Rubeis S, Lim ET, Sutcliffe JS, et al. Integrated model of de novo and inherited genetic variants yields greater power to identify risk genes. PLoS Genet. 2013;9(8):e1003671. pmid:23966865
  55. 55. Sanders SJ, He X, Willsey AJ, Ercan-Sencicek AG, Samocha KE, Cicek AE, et al. Insights into Autism Spectrum Disorder Genomic Architecture and Biology from 71 Risk Loci. Neuron. 2015;87(6):1215–33. pmid:26402605
  56. 56. Homsy J, Zaidi S, Shen Y, Ware JS, Samocha KE, Karczewski KJ, et al. De novo mutations in congenital heart disease with neurodevelopmental and other congenital anomalies. Science (New York, NY). 2015;350(6265):1262–6. pmid:26785492
  57. 57. Fromer M, Pocklington AJ, Kavanagh DH, Williams HJ, Dwyer S, Gormley P, et al. De novo mutations in schizophrenia implicate synaptic networks. Nature. 2014;506(7487):179–84. pmid:24463507
  58. 58. Xie Z, Bailey A, Kuleshov MV, Clarke DJB, Evangelista JE, Jenkins SL, et al. Gene Set Knowledge Discovery with Enrichr. Curr Protoc. 2021;1(3):e90. pmid:33780170
  59. 59. Kang HJ, Kawasawa YI, Cheng F, Zhu Y, Xu X, Li M, et al. Spatio-temporal transcriptome of the human brain. Nature. 2011;478(7370):483–9. pmid:22031440
  60. 60. Dougherty JD, Schmidt EF, Nakajima M, Heintz N. Analytical approaches to RNA profiling data for the identification of genes enriched in specific cells. Nucleic acids research. 2010;38(13):4218–30. pmid:20308160
  61. 61. Cantor R. Autism endophenotypes and quantitative trait loci. Autism Spectrum Disorders Oxford University Press: New York. 2011:690–704.
    • 62. Cantor RM, Navarro L, Won H, Walker RL, Lowe JK, Geschwind DH. ASD restricted and repetitive behaviors associated at 17q21.33: genes prioritized by expression in fetal brains. Molecular psychiatry. 2018;23(4):993–1000. pmid:28533516
    • 63. Rasmussen PB, Staller P. The KDM5 family of histone demethylases as targets in oncology drug discovery. Epigenomics. 2014;6(3):277–86. pmid:25111482
    • 64. Li X, Liu L, Yang S, Song N, Zhou X, Gao J, et al. Histone demethylase KDM5B is a key regulator of genome stability. Proceedings of the National Academy of Sciences. 2014;111(19):7096–101. pmid:24778210
    • 65. Kidder BL, Hu G, Zhao K. KDM5B focuses H3K4 methylation near promoters and enhancers during embryonic stem cell self-renewal and differentiation. Genome Biology. 2014;15(2):R32. pmid:24495580
    • 66. Zaidi S, Choi M, Wakimoto H, Ma L, Jiang J, Overton JD, et al. De novo mutations in histone modifying genes in congenital heart disease. Nature. 2013;498(7453):220–3. pmid:23665959
    • 67. Harrington J, Wheway G, Willaime-Morawek S, Gibson J, Walters ZS. Pathogenic KDM5B variants in the context of developmental disorders. Biochim Biophys Acta Gene Regul Mech. 2022;1865(5):194848. pmid:35905858
    • 68. Chen CY, Tian R, Ge T, Lam M, Sanchez-Andrade G, Singh T, et al. The impact of rare protein coding genetic variation on adult cognitive function. Nat Genet. 2023;55(6):927–38. pmid:37231097
    • 69. Huang Y, Bodnar D, Chen CY, Sanchez-Andrade G, Sanderson M, Shi J, et al. Rare genetic variants impact muscle strength. Nat Commun. 2023;14(1):3449. pmid:37301943
    • 70. Kaplanis J, Samocha KE, Wiel L, Zhang Z, Arvai KJ, Eberhardt RY, et al. Evidence for 28 genetic disorders discovered by combining healthcare and research data. Nature. 2020;586(7831):757–62. pmid:33057194
    • 71. Martin HC, Jones WD, McIntyre R, Sanchez-Andrade G, Sanderson M, Stephenson JD, et al. Quantifying the contribution of recessive coding variation to developmental disorders. Science (New York, NY). 2018;362(6419):1161–4. pmid:30409806
    • 72. Faundes V, Newman WG, Bernardini L, Canham N, Clayton-Smith J, Dallapiccola B, et al. Histone lysine methylases and demethylases in the landscape of human developmental disorders. The American Journal of Human Genetics. 2018;102(1):175–87. pmid:29276005
    • 73. Deciphering Developmental Disorders Study. Prevalence and architecture of de novo mutations in developmental disorders. Nature. 2017;542(7642):433–8. pmid:28135719
    • 74. Horton JR, Engstrom A, Zoeller EL, Liu X, Shanks JR, Zhang X, et al. Characterization of a Linked Jumonji Domain of the KDM5/JARID1 Family of Histone H3 Lysine 4 Demethylases. The Journal of biological chemistry. 2016;291(6):2631–46. pmid:26645689
    • 75. Dean SL, Tochen L, Augustine F, Ali SF, Crocetti D, Rajendran S, et al. The Role of the Cerebellum in Repetitive Behavior Across Species: Childhood Stereotypies and Deer Mice. Cerebellum. 2022;21(3):440–51. pmid:34390462