CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling

CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling | bioRxiv /* */ /* */ <!-- <!-- /*! * yepnope1.5.4 * (c) WTFPL, GPLv2 */ (function(a,b,c){function d(a){return"[object Function]"==o.call(a)}function e(a){return"string"==typeof a}function f(){}function g(a){return!a||"loaded"==a||"complete"==a||"uninitialized"==a}function h(){var a=p.shift();q=1,a?a.t?m(function(){("c"==a.t?B.injectCss:B.injectJs)(a.s,0,a.a,a.x,a.e,1)},0):(a(),h()):q=0}function i(a,c,d,e,f,i,j){function k(b){if(!o&&g(l.readyState)&&(u.r=o=1,!q&&h(),l.onload=l.onreadystatechange=null,b)){"img"!=a&&m(function(){t.removeChild(l)},50);for(var d in y[c])y[c].hasOwnProperty(d)&&y[c][d].onload()}}var j=j||B.errorTimeout,l=b.createElement(a),o=0,r=0,u={t:d,s:c,e:f,a:i,x:j};1===y[c]&&(r=1,y[c]=[]),"object"==a?l.data=c:(l.src=c,l.type=a),l.width=l.height="0",l.onerror=l.onload=l.onreadystatechange=function(){k.call(this,r)},p.splice(e,0,u),"img"!=a&&(r||2===y[c]?(t.insertBefore(l,s?null:n),m(k,j)):y[c].push(l))}function j(a,b,c,d,f){return q=0,b=b||"j",e(a)?i("c"==b?v:u,a,b,this.i++,c,d,f):(p.splice(this.i++,0,a),1==p.length&&h()),this}function k(){var a=B;return a.loader={load:j,i:0},a}var l=b.documentElement,m=a.setTimeout,n=b.getElementsByTagName("script")[0],o={}.toString,p=[],q=0,r="MozAppearance"in l.style,s=r&&!!b.createRange().compareNode,t=s?l:n.parentNode,l=a.opera&&"[object Opera]"==o.call(a.opera),l=!!b.attachEvent&&!l,u=r?"object":l?"script":"img",v=l?"script":u,w=Array.isArray||function(a){return"[object Array]"==o.call(a)},x=[],y={},z={timeout:function(a,b){return b.length&&(a.timeout=b[0]),a}},A,B;B=function(a){function b(a){var a=a.split("!"),b=x.length,c=a.pop(),d=a.length,c={url:c,origUrl:c,prefixes:a},e,f,g;for(f=0;f<d;f++)g=a[f].split("="),(e=z[g.shift()])&&(c=e(c,g));for(f=0;f<b;f++)c=x[f](c);return c}function g(a,e,f,g,h){var i=b(a),j=i.autoCallback;i.url.split(".").pop().split("?").shift(),i.bypass||(e&&(e=d(e)?e:e[a]||e[g]||e[a.split("/").pop().split("?")[0]]),i.instead?i.instead(a,e,f,g,h):(y[i.url]?i.noexec=!0:y[i.url]=1,f.load(i.url,i.forceCSS||!i.forceJS&&"css"==i.url.split(".").pop().split("?").shift()?"c":c,i.noexec,i.attrs,i.timeout),(d(e)||d(j))&&f.load(function(){k(),e&&e(i.origUrl,h,g),j&&j(i.origUrl,h,g),y[i.url]=2})))}function h(a,b){function c(a,c){if(a){if(e(a))c||(j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}),g(a,j,b,0,h);else if(Object(a)===a)for(n in m=function(){var b=0,c;for(c in a)a.hasOwnProperty(c)&&b++;return b}(),a)a.hasOwnProperty(n)&&(!c&&!--m&&(d(j)?j=function(){var a=[].slice.call(arguments);k.apply(this,a),l()}:j[n]=function(a){return function(){var b=[].slice.call(arguments);a&&a.apply(this,b),l()}}(k[n])),g(a[n],j,b,n,h))}else!c&&l()}var h=!!a.test,i=a.load||a.both,j=a.callback||f,k=j,l=a.complete||f,m,n;c(h?a.yep:a.nope,!!i),i&&c(i)}var i,j,l=this.yepnope.loader;if(e(a))g(a,0,l,0);else if(w(a))for(i=0;i (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0];var j=d.createElement(s);var dl=l!='dataLayer'?'&l='+l:'';j.src='//www.googletagmanager.com/gtm.js?id='+i+dl;j.type='text/javascript';j.async=true;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-M677548'); Skip to main content Home About Submit ALERTS / RSS Search for this keyword Advanced Search New Results CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling View ORCID Profile Mariana Moyses-Oliveira , Yating Liu , Serkan Erdin , View ORCID Profile Dadi Gao , Riya Bhavsar , Kiana Mohajeri , Kathryn O’Keefe , Philip M. Boone , Gabriela Xavier , Calwing Liao , Aiqun Li , Rachita Yadav , Monica Salani , Diane Lucente , Benjamin Currall , Celine E. F. de Esch , Derek J. C. Tai , Douglas Ruderfer , View ORCID Profile Kristen J. Brennand , James F. Gusella , View ORCID Profile Michael E. Talkowski doi: https://doi.org/10.1101/2025.10.27.684870 Mariana Moyses-Oliveira 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Mariana Moyses-Oliveira For correspondence: maricsmoyses{at}gmail.com mtalkowski{at}mgh.harvard.edu Yating Liu 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Serkan Erdin 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Dadi Gao 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Dadi Gao Riya Bhavsar 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Kiana Mohajeri 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA 5 Program in Biological and Biomedical Sciences, Harvard Medical School , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Kathryn O’Keefe 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Philip M. Boone 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA 6 Division of Genetics and Genomics, Boston Children’s Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Gabriela Xavier 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Calwing Liao 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA 7 Analytic and Translational Genetics Unit, Department of Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Aiqun Li 8 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai , New York, NY, USA 9 Nash Family Department of Neuroscience, Icahn School of Medicine at Mount Sinai , New York, NY, USA 10 Mount Sinai Center for Transformative Disease Modeling, Icahn School of Medicine at Mount Sinai , New York, NY, USA 11 Friedman Brain Institute, Icahn School of Medicine at Mount Sinai , New York, NY, USA 12 Icahn Institute for Data Science and Genomic Technology, Icahn School of Medicine at Mount Sinai , New York, NY, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Rachita Yadav 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Monica Salani 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Diane Lucente 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Benjamin Currall 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Celine E. F. de Esch 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Derek J. C. Tai 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Douglas Ruderfer 13 Division of Genetic Medicine, Department of Medicine, Vanderbilt Genetics Institute, Vanderbilt University Medical Center , 1211 Medical Center Dr. Nashville, TN, USA 14 Department of Biomedical Informatics and Department of Psychiatry and Behavioral Sciences, Vanderbilt University Medical Center , 1211 Medical Center Dr. Nashville, TN, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Kristen J. Brennand 8 Department of Genetics and Genomic Sciences, Icahn School of Medicine at Mount Sinai , New York, NY, USA 9 Nash Family Department of Neuroscience, Icahn School of Medicine at Mount Sinai , New York, NY, USA 10 Mount Sinai Center for Transformative Disease Modeling, Icahn School of Medicine at Mount Sinai , New York, NY, USA 11 Friedman Brain Institute, Icahn School of Medicine at Mount Sinai , New York, NY, USA 12 Icahn Institute for Data Science and Genomic Technology, Icahn School of Medicine at Mount Sinai , New York, NY, USA 15 Department of Psychiatry, Yale University New Haven , CT, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Kristen J. Brennand James F. Gusella 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA 5 Program in Biological and Biomedical Sciences, Harvard Medical School , Boston, MA, USA 16 Department of Genetics, Blavatnik Institute , Harvard Medical School, Boston, MA, USA 17 Harvard Stem Cell Institute, Harvard University , Cambridge, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site Michael E. Talkowski 1 Center for Genomic Medicine, Massachusetts General Hospital , Boston, MA, USA 2 Department of Neurology, Massachusetts General Hospital and Harvard Medical School , Boston, MA, USA 3 Stanley Center for Psychiatric Research, Broad Institute of MIT and Harvard , Cambridge, MA, USA 4 Program in Medical and Population Genetics, Broad Institute of MIT and Harvard , Cambridge, MA, USA 5 Program in Biological and Biomedical Sciences, Harvard Medical School , Boston, MA, USA Find this author on Google Scholar Find this author on PubMed Search for this author on this site ORCID record for Michael E. Talkowski For correspondence: maricsmoyses{at}gmail.com mtalkowski{at}mgh.harvard.edu Abstract Full Text Info/History Metrics Supplementary material Preview PDF ABSTRACT One of the seminal discoveries from genetic studies of autism spectrum disorder and related neurodevelopmental disorders (NDD) has been that loss-of-function (LoF) mutations in many genes that impact chromatin and transcriptional regulation confer substantial liability to NDD. Haploinsufficiency of the epigenetic regulator POGZ represents one of the strongest such associations; however, little is known about the direct or indirect regulatory targets of POGZ , or the mechanisms by which loss of this chromatin modifier alters early neuronal development and synaptic functions. Here, we created an allelic series of CRISPR-engineered human induced pluripotent stem cell (hiPSC) clones harboring mono- and biallelic POGZ deletions. In hiPSC-derived neural stem cells (NSC) and Neurogenin 2-induced neurons (iN), POGZ LoF altered the expression of genes associated with critical cellular processes and neuronal functions, including synaptic and intracellular signaling and extracellular matrix organization. Our multiomics profiling also showed altered footprinting of critical transcription factors (e.g., activator protein 1 complexes) that were enriched at promoters of differentially expressed genes associated with synaptic function. To further interrogate the shared molecular changes in neuronal development associated with NDD and POGZ regulation, we compared our results to deletions of the transcription factor MEF2C and the sodium channel gene SCN2A that we generated in these same isogenic iN. These analyses revealed strong enrichment of extracellular matrix and intracellular signaling disruption associated with POGZ and MEF2C deletion, whereas POGZ and SCN2A haploinsufficiency exhibited shared transcriptional effects on gene modules enriched for NDD-associated genes with opposing regulatory effects. Notably, we also observed alterations to synaptic firing rate and neurite extension with biallelic deletions, but not heterozygous lines, suggesting subtle effects in neuronal development associated with haploinsufficiency. Overall, these shared molecular consequences suggest key points of convergence that connect epigenetic regulation to neuronal function in the etiology of neurodevelopmental pathologies. INTRODUCTION Remarkable progress in recent years has identified shared genetic etiologies across a broad spectrum of neurodevelopmental disorders (NDD), including autism spectrum disorder (ASD), developmental delay, intellectual disability, and many diverse developmental syndromes 1 – 4 . Common hallmarks of these gene discovery efforts have been the consistently reproducible association of de novo loss-of-function (LoF) mutations in genes highly intolerant to functional variation in the general population (i.e., constrained) and directly involved in pathways governing synaptic activity and gene regulation (e.g., transcriptional regulation and chromatin organization) 1 , 5 – 9 . The primary challenge is now to connect the mechanisms by which alterations to genes involved in broad genome regulatory functions such as chromatin modification are interconnected, either directly or indirectly, with those genes involved in synaptic activity and collectively result in comparable deficits in neuronal function 9 – 11 . Some LoF mutations in chromatin-related genes can influence synaptic activity by dysregulating neuronal gene expression, or through other mechanisms such as proximal interactions, post-translational modifications, protein-protein interactions, or signaling pathways 7 , 9 , 11 – 17 . However, the broader convergence of chromatin and synaptic mechanisms, as well as their involvement in neurodevelopment and NDD pathogenesis remain uncertain. POGZ (Pogo transposable element derived with ZNF domain) is among the chromatin-related genes with the most robust statistical evidence for association with ASD and more broadly defined NDD 1 , 7 . Individuals with heterozygous LoF mutations in POGZ often present with ASD, intellectual disability (ID), and/or White-Sutton syndrome 18 , 19 . Mouse models with homozygous Pogz LoF or expression of a POGZ patient-specific missense mutation show impairments in neurogenesis, neurodifferentiation, neuronal migration, and electrophysiological activity 20 , 21 . However, little is known concerning POGZ’s role in human neurodevelopment. Biochemical studies have shown that POGZ forms complexes with other ASD/NDD-associated chromatin regulators, e.g., ADNP, CHD4, and the BAF complex 22 , 23 . Also, the POGZ-interacting protein, heterochromatin protein 1 (HP1), is known to modulate chromatin topology in pericentromeric heterochromatin regions and euchromatic sites 24 , 25 . Although POGZ was first described as a transcriptional repressor through its interaction with HP1 26 , 27 , it has been suggested recently that POGZ might modulate gene expression as both a repressor and an activator 21 , 22 . Consequently, the impact of POGZ disruption on neuronal function and potential convergence with the disruption of genes involved in synaptic activity are fundamental insights that are required to gain understand the POGZ-associated neurodevelopmental syndromes and traits. We took a systematic neuronal modeling approach to interrogate the impact of POGZ haploinsufficiency on cellular functions in developing neural cells. We engineered an allelic series of isogenic human induced pluripotent stem cell (hiPSC)-derived neural stem cell (NSC) and Neurogenin 2 ( Ngn2 )-induced neuron (iN) models that harbored CRISPR generated deletions of POGZ. We used heterozygous lines to explore the transcriptional, regulatory, and cellular consequences of POGZ haploinsufficiency. Additional lines with homozygous POGZ knockout grounded validation and mechanistic analysis of POGZ function. We then compared transcriptional signatures of POGZ LoF to the effects of deletions engineered on the same isogenic backgrounds for two well-established ASD/NDD genes associated with POGZ regulatory changes — MEF2C , which encodes a transcriptional regulator, and SCN2A , a sodium channel gene. Our analyses disclosed functional signatures implicating direct and indirect POGZ regulatory changes that altered transcription factor binding at the promoters of synaptic proteins and shared molecular consequences with MEF2C and SCN2A that converged at the level of genes associated with altered neuronal development. RESULTS Generation of an allelic series of isogenic neural cell lines harboring POGZ deletions We created POGZ CRISPR-mediated LoF models on two different hiPSC backgrounds (GM08830 [male] and MGH2069 [female]), including a full and a partial deletion of POGZ , and corresponding unedited wild-type (WT) models that were exposed to the same CRISPR guide RNAs and single-cell sorting ( Figure 1A , Supplementary Figure 1A and 1B ). We differentiated these lines into two distinct neural cell types relevant to neuronal development and function, i.e., NSC and iN ( Figure 1B ), and evaluated transcriptional profiles by RNA sequencing (RNA-seq) and chromatin accessibility by the assay for transposase-accessible chromatin using sequencing (ATAC-seq) ( Figure 1C ). We generated a total of 66 RNA-seq libraries plus 66 ATAC-seq libraries for each cell type (NSC and iN), derived from 66 hiPSC differentiation replicates. The differentiation replicates were originated from a total of 23 and 24 independently CRISPR-engineered clonal hiPSC lines clones in GM08830 and MGH2069 backgrounds, respectively. We performed differential expression analyses after validating POGZ dosage and downregulation of RNA and protein expression in the edited lines ( Supplementary Figure 1C-F ). Comparison of the heterozygous POGZ LoF lines to the matched unedited isogenic control lines discovered differentially expressed genes (DEG) resulting from POGZ haploinsufficiency. In contrast, lines with complete loss of POGZ due to compound heterozygous LoF mutations provided validation and further mechanistic insights. In all experiments, we assessed DEG for biological significance using a false discovery rate (FDR) q<0.1 threshold and concordant direction of effect across two different hiPSC backgrounds (e.g., consistently upregulated or downregulated DEG in GM08830 and MGH2069 backgrounds; see Supplemental information for details). Download figure Open in new tab Figure 1. Characterization of the consequences of POGZ loss of function in neuronal models. (A) CRISPR-engineering of whole gene deletion (del 1) and partial gene deletion (del 2) of POGZ gene in two different backgrounds of human induced pluripotent stem cells (hiPSC) (GM08830 [Donor A, male] and MGH2069 [Donor B, female]), resulting in heterozygous (het) and compound heterozygous (comp het) clones. (B) Neuronal differentiation of hiPSC clones into neural stem cells (NSC) and Ngn2-induced neurons (iN). (C) Molecular profiling of hiPSC-derived iN and NSC using RNA-seq to identify differentially expressed genes (DEG) and ATAC-seq to address differentially accessible regions (DARs). (D) Cellular phenotyping of iN with multielectrode array to measure firing activity and incucyte to measure neurite length. POGZ haploinsufficiency is associated with altered expression of genes related to extracellular matrix and synaptic processes We first focused on heterozygous POGZ deletions to gain insight into NDD-associated POGZ haploinsufficiency. Neuronal lines differentiated from the different hiPSC backgrounds showed significant directionally-concordant overlap of DEG across the backgrounds (GM08830 and MGH2069) in the transcriptional effects of heterozygous POGZ deletion ( Supplementary Figure 2A-D ). In heterozygous iN, we observed 424 DEG (146 upregulated and 278 downregulated; Figure 2A , Supplementary Table 1 ) that were significantly enriched for molecular pathways related to synapse activity and extracellular matrix, calcium signaling, and cell migration, among others ( Figure 2B ). In heterozygous NSC, the 1,115 DEG (664 upregulated and 451 downregulated; Figure 2A , Supplementary Table 2 ) were significantly enriched for similar extracellular matrix and focal adhesion-related pathways ( Figure 2B ). There was also robust statistical evidence for overlap between the DEG in iN and NSC (Fisher’s exact test, p-value=1.14e-10; Supplementary Figure 2E and F ). Most significantly enriched pathways among iN and NSC DEG comprised a mixture of downregulated and upregulated genes ( Supplementary Figure 3A ), suggesting that the functional consequences of POGZ perturbation involve both transcriptional activation and repression. Download figure Open in new tab Figure 2. Transcriptional profiles of NSC and iN lines with heterozygous POGZ deletion (A) Number of upregulated (blue) and downregulated (red) genes in iN and NSC heterozygous (het) models. (B) Statistical significance for enrichment of pathways among DEG in heterozygous (het) N and NSC lines. Numbers in parentheses indicate the total number of genes in a given gene list. Colors represent pathway categorization according to the nature of the functional categories related to the enriched term. Groups of functional categories: VESIC: vesicles, ENERGY: energy metabolism, MENTAL: mental and behavioral traits, IMMUNE: immune system and inflammation, MIG: cell migration, CALCIUM: calcium signaling, MIT: mitosis and cell proliferation, STEROID: steroid metabolism, CSKEL: cytoskeleton, CHROM: chromatin structure and regulation, ECM: extracellular matrix, SYN: synapse. C. Heatmap indicating significant overlap between co-expression modules in iN and NSC. Asterisks indicate the modules containing POGZ. Next, we established modules of co-expressed genes in NSC and iN by using weighted gene co-expression network analysis (WGCNA) 28 . For each cell type, we merged the data across both backgrounds and all deletions models (see Supplemental information for details). In WGCNA of heterozygous and control iN, six co-expression modules were significantly correlated with POGZ genotype (Bonferroni-adjusted p-value<0.1); three displayed downregulation and three upregulation of the module eigengene ( Supplementary Figure 3B ). In the heterozygous and control NSC, four co-expression modules were significantly correlated with POGZ genotype (Bonferroni-adjusted p-value<0.1), with two having eigengenes indicative of downregulation and upregulation, respectively ( Supplementary Figure 3C ). All DEG and genotype-correlated co-expression modules in both NSC and iN were significantly enriched for DEG associated with Pogz dosage in previously published mouse models ( Supplementary Figure 4A and B ), indicating reproducibility of POGZ LoF impacts across species. POGZ was a member of the downregulated modules “red” (619 genes) and “blue” (1,496 genes) in iN and NSC, respectively ( Supplementary Figure 3B and C, Supplementary Table 3 ). The pathway enrichment analysis of these two modules showed a strong involvement of POGZ co-expressed genes in cell type-specific functional categories. The iN red module was significantly enriched for processes related to extracellular matrix and growth factor binding (FDR=1.50e-6, FDR=4.43e-2, respectively), and the NSC blue module was significantly enriched for pathways associated with translation, ribosomal RNA processing and nervous system development (FDR=1.83e-168, FDR=2.18e-61, FDR=2.52e-30, respectively) ( Supplementary Figure 3D ). These modules shared 46 genes, an overlap that did not reach significance (p-value=1.93e-1, Figure 2C ); however, the POGZ- containing downregulated iN red module showed significant overlap with NSC upregulated modules, “turquoise” (overlap of 85 genes, p-value=1.22e-5) and “brown” (overlap of 51 genes, p-value=2.03e-3) ( Supplementary Table 4, Figure 2C ). These gene overlaps were significantly enriched for enzyme-receptor signaling, growth factor binding, extracellular matrix organization and neurodevelopmental pathways that were also captured in the POGZ -containing modules ( Supplementary Figure 3D ). Thus, these transcriptional profiles indicate that POGZ LoF results in gene expression disturbances that can be cell-type specific or shared but directionally discordant yet converge on common cellular processes. POGZ direct regulatory targets modulated in edited NSC are associated with chromatin regulation, cell cycle, and RNA processing Gene expression differences in our neuronal models could be due to direct transcriptional regulation of DEG by POGZ binding at the promoter (see definitions of POGZ direct effects in Supplementary Figure 5A ). To interrogate the modulation of genes that are direct targets of POGZ, we compared the iN and NSC DEG with POGZ targets from published CUT&RUN or CUT&Tag experiments defining POGZ binding sites in human fetal cortex 22 (human POGZ targets) and mouse embryonic stem cells (mESC) 23 (mouse Pogz targets), respectively ( Supplementary Table 5 ). Downregulated DEG from heterozygous NSC lines overlapped significantly with the human POGZ targets (p-value=1.06e-4), and both the upregulated and downregulated NSC DEG showed mild enrichment of mouse Pogz targets ( Supplementary Figure 6A ). The smaller number of DEG from heterozygous iN lines showed no such enrichments ( Supplementary Figure 6A) . Across the WGCNA co-expression modules correlated with POGZ genotype, only the downregulated NSC blue module containing POGZ showed strong enrichment of POGZ target genes ( Supplementary Figure 6B ). These 498 overlapping genes were significantly enriched for DNA binding, transcription, and RNA processing-associated processes ( Supplementary Figure 6C ). The same pathways were also detected by the 145 genes shared between NSC downregulated DEG and human POGZ targets ( Supplementary Figure 6C ), arguing that one disrupted function in NSC harboring POGZ haploinsufficiency was the role of activating expression of some direct target genes involved in these biological processes. However, these analyses also argue more broadly that most gene expression changes caused by POGZ heterozygous disruption do not result from direct regulation by POGZ binding at the promoter of the DEG, but rather indirect regulatory changes. Complete inactivation of POGZ impacts activity of its direct target genes We next sought to confirm these new insights into POGZ function by investigating cell lines with complete inactivation due to compound heterozygous mutations. The strong transcriptional consequences (9,006 and 5,148 DEG in iN and NSC, respectively, Supplementary Figure 7A, Supplementary Table 6 and 7 ) were highly correlated with the impacts of heterozygous LoF mutations ( Supplementary Figure 7B-F ). Similarly to our findings for iN and NSC co-expression modules, subsets of DEG overlapped between different zygosities and cell types ( Supplementary Figure 7B and C ). Some of these DEG overlaps presented with discordant directions (e.g., downregulated in heterozygous iN and upregulated in compound heterozygous NSC) ( Supplementary Figure 7B ). In compound heterozygous iN, POGZ direct targets ( Supplementary Figure 7G ) overlapped significantly with upregulated DEG (p-value=1.48e-23), implicating POGZ as a transcriptional repressor in this cell type. These POGZ direct target DEG overlapped significantly with NDD genes previously implicated by large-scale exome studies ( Supplementary Figure 7H ), but this pattern was not observed in the context of heterozygous mutations ( Supplementary Figure 7I ). The NDD-associated direct POGZ target genes upregulated in iN showed significant enrichment for pathways related to synapse, immune, cell cycle, and translation regulation processes ( Figure 3 ). Similar functional categories were recapitulated by the pathways enriched among NDD-associated direct POGZ target genes downregulated in NSC ( Figure 3 ). These analyses indicate that POGZ directly modulates the transcription of other NDD-associated genes. Download figure Open in new tab Figure 3. Impacts of compound heterozygous POGZ deletion on ASD and NDD-associated genes Statistical significance for enrichment of pathways among NDD-associated genes that are downregulated (left panel) or upregulated (middle panel) differentially expressed POGZ targets in iN compound heterozygous lines. Statistical significance for enrichment of pathways among NDD-associated genes that are downregulated differentially expressed POGZ targets in NSC compound heterozygous lines (right panel). Intersection of two gene lists is represented by “⋂”. Numbers in parentheses indicate the total number of genes in a given gene list. Colors represent pathway categorization according to the nature of the functional categories related to the enriched term. Groups of functional categories: APO: apoptosis, VESIC: vesicles, VASC: vascular system, ENERGY: energy metabolism, IMMUNE: immune system and inflammation, MIG: cell migration, MIT: mitosis and cell proliferation, TRANSLATION: translation, RNA: RNA metabolism, STEROID: steroid metabolism, CSKEL: cytoskeleton, CHROM: chromatin structure and regulation, SIG: intracellular signaling pathways, ECM: extracellular matrix, SYN: synapse. BH: Benjamini-Hochberg adjusted p-values. POGZ’s interactors influence the direction of its transcriptional modulation One potential explanation for cell type-dependency of directional differences in the regulation of POGZ targets pertains to the protein(s) with which POGZ is complexed when it binds to DNA. POGZ is known to form a variety of complexes with chromatin proteins, such as HP1, as well as other NDD risk-associated transcription regulators, such as ADNP, CHD4, and BRG1, the latter being a member of the BAF complex, which has also been implicated in NDD 21 – 23 , 29 . To investigate the influence of POGZ’s interactors on expression modulation, we compared NSC and iN DEG with lists of mouse Pogz targets defined with respect to the composition of the POGZ-containing protein complex (See methods for details) 23 . Importantly, all such subsets of mouse Pogz targets showed significant overlap with POGZ targets in the human fetal cortex ( Supplementary Figure 8A ), and the mono- and biallelically edited lines demonstrated concordant DEG directionalities in the context of POGZ complex composition ( Supplementary Figure 9 ). As expected, compound heterozygous iN lines revealed a strong correlation between upregulated DEG and genomic targets of multiple POGZ complex configurations ( Supplementary Figure 8B ), reinforcing POGZ’s important role as a repressor of gene expression in these cells, though a significant subset of Pogz/Chd4 targets were downregulated, consistent with an activator role. In NSC compound heterozygous lines, genes matching mouse Pogz/ Chd4 targets were significantly enriched for both up and downregulated DEG ( Supplementary Figure 8B , p-value=1.88e-3 and 2.91e-3 for up and downregulated DEG, respectively), supporting dual directionality of modulation by these complexes. However, the presence/absence of ADNP in the POGZ-containing complex showed mirrored effects on gene expression: upregulated DEG were significantly enriched for mouse Pogz/Adnp targets while downregulated DEG were significantly enriched for mouse Pogz targets of complexes without Adnp ( Supplementary Figure 8B ). Comparison of the complex-informed subsets of POGZ targets with Pogz LoF DEG across different mouse brain regions further emphasized that the direction of modulation of POGZ-regulated genes can vary according to the tissue investigated ( Supplementary Figure 10 ). For example, Pogz/Chd4 target genes were significantly enriched among the downregulated mouse cerebellum DEG, yet among the upregulated hippocampus DEG and human compound het iN DEG ( Supplementary Figure 8B, Supplementary Figure 10 ). Additional differences in pattern by brain region were evident for other Pogz complex targets indicating that the direction of effect on POGZ direct regulatory targets varies across cell type and possibly species, presenting a complexity that will likely require future single-cell analyses to resolve. POGZ haploinsufficiency impacts gene regulation of non-POGZ targets In some instances, transcription factors (TFs) can produce nucleosome repositioning through the recruitment of chromatin remodelers to promote changes in DNA accessibility. To address a POGZ regulatory impact on chromatin accessibility in the disease-relevant genotypic context, we examined ATAC-seq profiles in the heterozygous cell models. We first identified differential accessibility regions (DARs) comparing WT and heterozygous deletion lines ( Supplementary Table 8) , then investigated DARs enrichment among promoters of DEG. We discovered that DEG in both iN and NSC displayed a strong association with DARs in these neuronal models, emphasizing the value of such multiomic profiling ( Supplementary Figure 11 , p-value=9.11e-10 and 7.14e-3 for iN and NSC, respectively). In heterozygous lines, there was no association between POGZ target DEG and chromatin accessibility changes at gene promoters, but we did observe strong association in homozygous models ( Supplemental material ). Additionally, most of the DEG observed in the context of POGZ haploinsufficiency were not genes known to be direct POGZ targets ( Supplementary Figure 6A ). This evidence suggests that DARs and DEG in heterozygous lines might be driven by indirect regulatory mechanisms. Therefore, we defined POGZ’s indirect effects as the impact of other TFs over DEG in heterozygous deletion lines. We hypothesize that these indirect effects might occur due to differential expression of other TFs themselves or due to differential accessibility where other TFs bind (see definitions of indirect effects on Supplementary Figure 5B ). We then leveraged another layer of ATAC-seq data to explore this hypothesis. Indirect effects are mediated by other transcription factors To gain molecular insight into POGZ’s indirect effects, we profiled accessible binding sites for a broad range of TFs in the haploinsufficiency models. From deep coverage ATAC-seq analysis, we called the “transcription factor footprint”, considered as a proxy for TF binding probability, based on the depletion of ATAC-seq local signal over a TF motif within an open chromatin region 30 (see Methods for details). The aim of this approach was to reveal those TF whose binding was most susceptible to POGZ haploinsufficiency, potentially reflecting indirect POGZ regulation. We dissected the footprints for 429 and 393 motifs, whose corresponding TFs were expressed in iN and NSC models (based on our RNA-seq datasets), respectively ( Supplementary Table 9 ). Briefly, for each TF motif we first scored the genome-wide footprint differences between genotypes by summarizing the changes at all of its matching loci. When any motif of a single TF genome-wide change was significant (See Methods), we considered it a differentially footprinted TF (DTF) and further screened its footprint at each motif-containing locus. In iN, the 16 DTFs identified from genome-wide scoring all indicated reduced footprinting in POGZ heterozygous LoF cells. Of 39 DTFs in heterozygous NSC, 20 and 19 indicated increased and reduced footprinting, respectively ( Supplementary Table 9 ). In iN, 31 (∼2.5%) of 1,245 expressed TFs are DEG and none is a DTF, while in NSC, 89 (∼7.6%) out of 1,168 expressed TFs are DEG and only one, SNAI2 , is also a DTF. We further confirmed that the potential binding of SNAI2 was not enriched in the promoter of NSC DEG. There were 14 DTFs shared by both cell types and many of these involve related protein complexes. Together they comprise proteins encoded by 11 genes, none of which was a DEG in our heterozygous cellular models ( Supplementary Figure 12A ). The subset of genes whose promoters contain the DTF binding sites (i.e., DTF targets) overlapped significantly with DEG ( Figure 4A and B ). The pathways strongly enriched among differentially expressed POGZ targets ( Supplementary Figure 6C ) were not significantly enriched among differentially expressed DTF targets ( Figure 4C ). Thus, in haploinsufficient POGZ models the downregulation of genes related to DNA binding, transcription and RNA processing ( Supplementary Figure 6C ) is specifically caused by POGZ direct regulation not indirect regulatory effects ( Figure 4C ). Download figure Open in new tab Figure 4. Integration between differential expression and footprinting analysis to identify indirect regulatory mechanisms (A) and (B) Overlap between DEG and DTF targets from human heterozygous (het) iN or NSC. (C) Statistical significance for enrichment of pathways among heterozygous iN (left panel) and NSC (right panel) DTF targets which are DEG. Intersection of two gene lists is represented by “⋂”. Numbers in parentheses represent total genes in a given list. Colors represent pathway categorization according to the nature of the functional categories related to the enriched terms. Groups of functional categories: APO: apoptosis, IMMUNE: immune system and inflammation, MIG: cell migration, MIT: mitosis and cell proliferation, STEROID: steroid metabolism, SIG: intracellular signaling pathways, ECM: extracellular matrix, SYN: synapse, VASC: vascular. (D) and (E) FOS-JUN footprinting score over gene promoters of down/upregulated FOS-JUN targets in heterozygous iN (D) or NSC (E) lines. Numbers in parentheses indicate the total number of genes in a given gene list. BH: Benjamini-Hochberg adjusted p-values. We then sought to ascertain whether any DTF shared between cell types might represent a common element of POGZ regulation. Of 14 DTFs found in both cell types ( Supplementary Table 9 ), 13 belong to a well-known immediate early gene transcriptional regulator 31 , 32 , the AP1 (activator protein-1) complex family, with JUN and/or FOS components ( Supplementary Table 9 ). In agreement with the globally increased footprinting of AP1 complexes in NSC and decreased footprinting in iN, AP1 regulatory targets among DEG were mainly upregulated by heterozygous POGZ LoF in NSC and downregulated in iN ( Figure 4D and E ). In both cell types, the fold changes of footprinting at the AP1 complex binding sites were in the same direction as the expression of their target genes ( Figure 4D and E ), in accordance with the transcriptional activator activity previously described for AP1 complexes 31 , 32 . Because neither this nor any other class of TFs was enriched among overall DEG, it is the activity of this class of TFs that is modulated by POGZ haploinsufficiency. Their impact largely differs in direction between NSC and iN models, but accounts for a substantial portion of the altered gene expression in NSC and iN. Neuronal haploinsufficiency of two NDD-associated chromatin genes shows molecular pathway convergence Chromatin remodelers and transcriptional regulators constitute functional classes consistently found to be disrupted in large-scale exome sequencing studies of ASD/NDD 1 , 7 , 13 . To identify shared molecular signatures by which chromatin-related NDD-risk genes might contribute to pathology, we compared the transcriptional profiles from independent neuronal haploinsufficiency models for POGZ and MEF2C , another well-established NDD-associated gene, encoding Myocyte Enhancer Factor 2C 33 . MEF2C was also found to be a DTF target in the POGZ heterozygous deletion NSC. The hiPSC-derived neural models (iN and NSC) for both gene perturbations were generated strictly using the same CRISPR editing and differentiation protocols 33 . Since the MEF2C models were created on the GM08830 hiPSC background, we focused on the POGZ GM08830 iN and NSC cells, assuring a direct isogenic comparison of the RNA-seq datasets not only between conditions (i.e., edited vs. unedited clones) but also between genes (i.e., POGZ vs. MEF2C ). There was a significant overlap of DEG at FDR<0.1 between these haploinsufficiency models ( Figure 5A ). Shared DEG in iN were significantly enriched for processes related to intracellular signaling and extracellular matrix organization ( Figure 5C ), while in NSC, they were related to chromatin and transcription regulation, including gene targets of AP-1 complexes ( Supplementary Figure 13 ). Download figure Open in new tab Figure 5. Overlap of transcriptional profiles derived from isogenic heterozygous iN with POGZ, MEF2C or SCN2A gene deletions (A) Heatmap indicating statistical significance for overlap between DEG (FDR<0.1) from heterozygous iN with POGZ, MEF2C, or SCN2A gene deletions. (B) Statistical significance for enrichment of pathways among DEG overlaps from two-way comparisons of heterozygous iN with POGZ, MEF2C or SCN2A gene deletions. Numbers in parentheses indicate the total number of genes in a given gene list. Colors represent pathway categorization according to the nature of the functional categories related to the enriched term. Groups of functional categories: IMMUNE: immune system and inflammation, MIG: cell migration, MIT: mitosis and cell proliferation, STEROID: steroid metabolism, SIG: intracellular signaling pathways, ECM: extracellular matrix. BH: Benjamini-Hochberg adjusted p-values. Numbers in parentheses indicate the total number of genes in a given gene list. Similarities and differences due to haploinsufficiency of NDD-associated chromatin and synapse genes in human neurons We next expanded our functional analyses to compare our POGZ LoF models with LoF of another NDD-associated gene with a different biological role: SCN2A encoding the alpha 2 subunit of the voltage-gated sodium channel, thought to cause NDD through its effect on synaptic activity, is among the most robustly associated ASD genes 1 . Since SCN2A is expressed at very low levels in NSC, we only focused this comparison on iN edited lines generated on the same isogenic background as those for POGZ using similar procedures. We performed WGCNA co-expression analysis considering heterozygous mutant iN models. Both heterozygous deletion and control iN samples for POGZ (23 samples) and SCN2A (35 samples) from GM80830 background were considered, whereas MEF2C iN samples (11 samples) were excluded as they did not meet the criteria for the minimum number of clones (see Supplemental Information for details). The “green” module (N=647) presented discordant regulation between POGZ and SCN2A edited lines (upregulated and downregulated, respectively Supplementary Table 10 ). It was strongly enriched for ASD/NDD associated genes and LoF constrained genes and mildly enriched for chromatin modifiers, but not synaptic genes ( Supplementary Figure 14, Supplementary Figure 15 ). This suggested that POGZ and SCN2A haploinsufficiency can impact some of the same neurodevelopmental processes, but in opposite directions. The “tan” (N=395) and “purple” (N=603) modules also reflected opposing transcriptional effects but without disease-associated enrichments. Only in the “midnightblue” module (N=341) was the direction of effect the same (downregulation) for both POGZ and SCN2A haploinsufficiency. Together, these analyses suggested a mild convergent signal between POGZ and SCN2A haploinsufficiency involving co-modulation of NDD-associated and constraint genes, but with transcriptional changes in opposite directions. We next employed microelectrode array (MEA) for electrophysiological comparison of the isogenic iN with heterozygous POGZ or SCN2A gene deletions ( Figure 1D ). The mild WGCNA-defined differences did not translate into alterations in firing rate, where subtle effects did not achieve statistical significance. In contrast, when both alleles were edited to achieve complete LoF, both POGZ and SCN2A models displayed dramatic effects on neuronal firing rate, but in opposite directions, with increased firing in POGZ compound heterozygous iN and decreased firing in homozygous SCN2A iN ( Figure 6A and B ). Changes in neurite length, measured using label-free Incucyte, confirmed the strong impact of the total functional inactivation of these genes in iN ( Figure 6C and D ), with POGZ compound heterozygous iN exhibiting a more intricate neuronal network and homozygous SCN2A lines showing reduced neurite length and branching ( Figure 6 ). Download figure Open in new tab Figure 6. Cellular phenotyping on isogenic heterozygous iN with POGZ or SCN2A gene deletions (A) and (B) . Weighted mean firing rate at the time point with highest activity measured by multi-electrode array in iN with POGZ heterozygous (het) and compound heterozygous (comp het) deletions (A) and with SCN2A het and homozygous (hom) deletions (B). (C) and (D) . Neurite length normalized by cell body cluster area at the time point with highest extension density measured by Incucyte in iN with POGZ heterozygous (het) and compound heterozygous (comp het) deletions (C) and with SCN2A het and homozygous (hom) deletions (D). Asterisk indicates p-value<0.01. In conclusion, while heterozygous LoF of POGZ and SCN2A exhibited subtle and often opposing transcriptional effects, particularly within modules that were significantly enriched for NDD-associated genes, these changes did not translate to significant alterations in neuronal firing rate or neurite extension in iN. However, the complete, biallelic LoF of these genes revealed robust yet contrasting impacts on neuronal function, arguing that while both genes contribute to NDD their specific mechanisms and the direction of their impact might differ. DISCUSSION Unlike functional studies using cell lines derived from patients, introducing genomic variants of interest into an isogenic background permits direct comparisons across genes and mutations for a broader scope of evaluation of regulatory mechanisms, including efforts to infer causal relationships between variant, expression profiles, chromatin accessibility regulation, and neuronal phenotypes 34 – 36 . Our CRISPR engineering experiments yielded a catalog of 66 hiPSC replicates, with different zygosities across two different background lines. The integration of sequencing datasets from this compendium of models allowed us to statistically dissect the molecular signals which are consistent between multiple experiments. The significant overlap between the biological signatures observed in models derived from different hiPSC backgrounds indicated that the impacts due to POGZ LoF described here are reproducible. The comparison across different cell types allowed a broader scope of evaluation of regulatory mechanisms. It has been demonstrated that the expression of chromatin-related NDD genes peaks in early neurodevelopment (8-9 human gestational weeks) 37 , 38 and, when detected in later stages, is enriched in cortical layers, including excitatory neuronal lineages 39 . Additionally, broader genetic studies in neuropsychiatric diseases 40 and single-cell transcriptomics in ASD 41 have implicated neuronal genes that converge on cellular programs associated with excitatory neurons. Thus, the targeted cell types investigated here – NSC and iN – correspond to disease-relevant contexts at early developmental stages and mature synaptic functioning, respectively. Functional impacts due to POGZ LoF frequently affected similar gene sets in NSC and iN; however, in many instances, the directions of effect were opposite in the two cell types. These differences were partially related to the specific composition of POGZ protein complexes, and the gene targets directly regulated by those different versions. In agreement with previous POGZ functional investigations 22 , POGZ heterozygous models presented with subtle molecular impacts, especially regarding transcriptional disturbances of POGZ target genes. In heterozygous models, these transcriptional disturbances are partially derived from indirect regulatory effects (i.e, transcriptional disturbance of DTF target genes), and AP1 complexes are the main class of TFs with regulatory function affected in both cell types. AP1 complexes are long-standing immediate early gene regulators, with drastic implications for synaptic activity 42 . Mouse knockouts for Actl6B , a BAF complex component, have previously shown impairment in early gene response through disruption of AP1 complex regulatory activity 43 . Additionally, increased c-fos protein expression associated with higher excitatory synaptic inputs have been observed in the mouse brain with the LoF of Auts2 , a well-known ASD-associated gene 44 . Beyond POGZ -specific functional impacts, this study took advantage of CRISPR-engineered isogenic neuronal heterozygous models to explore the molecular convergence between POGZ and other ASD/ NDD benchmark genes from different functional classes: SCN2A , the top ASD-associated synaptic gene, and MEF2C , which is one of the main NDD-associated TFs and was also found to be a DTF target in lines with POGZ heterozygous deletion. POGZ and MEF2C showed strong convergence of their molecular impacts in NSC and iN, with substantial overlaps on overall transcriptional disturbance and affected biological pathways, including disruption of AP1 complexes regulation and extracellular matrix. Previous MEA studies performed by our research group applied analogous functional protocols on isogenic neuronal lines with MEF2C LoF 33 . When the results from this study are compared to the MEA data presented here, we can conclude that LoF of POGZ and MEF2C show opposite effects on firing activity, with POGZ LoF models showing increased firing rate, and the isogenic models of MEF2C LoF showing decreased firing rate. These results suggest that despite the striking transcriptional convergence between MEF2C and POGZ iN lines, shared transcriptional signals between different NDD/ASD genes might be translated into effects on the same cellular processes, but in different directions. Co-expression networks in heterozygous iN showed that constrained NDD-associated genes were modulated in opposite direction in POGZ and SCN2A models, with inverse functional consequences in the cellular assessments between these gene perturbations. The decrease in neuronal excitability in human iN with biallelic mutations in SCN2A is concordant with previous studies 34 , while the increased excitability in POGZ models for POGZ biallelic mutations is unprecedented in the literature and might be specific for the cell type here evaluated. Although the complete removal of POGZ or SCN2A do not reflect the condition in human disease, the biallelic models generated in this study provided mechanistic insights associated with neuronal abnormalities. Alterations in the excitation-inhibition balance ratio in cerebral cortex is a common feature in animal models of rare genetic variants associated with ASD and have been proposed as a relevant pathophysiological mechanism in ASD 45 – 47 . The altered excitability observed in our knockout POGZ and SCN2A models might be relevant to understanding the cellular consequences of these genomic perturbations in the in vitro neuronal used in this study. Our study design presents some limitations. Monolayer cellular cultures fail to capture the cellular complexity of the human brain. Although NSC and iN can be considered as representative of different developmental stages, they are not appropriate models for trajectories of cellular differentiation. This study considered POGZ direct regulatory targets defined in previously published functional experiments which were based on human fetal cortex and mESC 22 , 23 . POGZ binding sites in iN and NSC might differ; however, the overall profile is expected to be conserved across cell types, given the strong overlap observed between human fetal cortex and mESC POGZ targets. These limitations indicate that our results should be interpreted cautiously with respect to clinical implications. It is important to note that more than half of DEG were neither POGZ known targets nor DTF targets, and other regulatory mechanisms remain to be identified. Nevertheless, by addressing the effect of the interrogated gene disruptions on specific neuronal cell types with isogenic backgrounds, we were able to dissect the functional assessment of isolated cell populations of great interest in NDD pathogenesis, resulting in unambiguous data interpretation. We present CRISPR-engineered isogenic allelic series of POGZ disruption in human-derived in vitro models and functional characterization of resultant transcriptional and chromatin accessibility effects. These studies uniquely dissect the parallel consequences of POGZ direct and indirect regulatory effects on heterozygous cell lines, which carry the disease-relevant genotypic context. Our study clearly demonstrates convergence between ASD/ NDD genes and highlights commonly affected pathways, which might be pharmacologically targeted in future investigations. METHODS Engineering of CRISPR-mediated gene deletions Human induced pluripotent stem cell (hiPSC) lines were maintained on Matrigel-coated dishes (Corning) with Essential 8 culture medium (Thermo Fisher Scientific) and incubated at 37°C in a humidified atmosphere with 5% CO2. CRISPR manipulations on POGZ locus were performed in two hiPSC lines with independent genetic backgrounds, using one male (GM08830 48 ) and one female line (MGH2069 49 ). Those two hiPSC lines were previously deep whole genome sequenced to confirm the absence of NDD-associated variants. Characterization of pluripotency has been previously described 50 . Two POGZ deletion types (one full and one partial gene deletion) were attempted to increase the probability of editing efficiency. Both deletion types are intended to cause POGZ LoF. Cas9 guide RNA (gRNA) sequences were designed immediately 5′ and 3′ to POGZ gene and at POGZ intron 12 (NM_015100, gRNA sequences are available in Supplementary Table 11 ) and obtained as modified synthetic sgRNA (Synthego). hiPSCs were co-transfected with the ribonucleoprotein (RNP) complex [Cas9 (Alt-R® S.p. HiFi Cas9 Nuclease V3, Integrated DNA Technologies)+gRNA] and CleanCap EGFP mRNA (TriLink Biotechnologies) using Lipofectamine Stem Transfection Reagent (Invitrogen) according to manufacturer’s instructions. To generate the whole gene deletion (del1, 56kb) and partial gene deletion (del2, 7kb) in heterozygous state, wild type hiPSCs were transfected with two different combinations of gRNAs: (1) 5’ plus 3’ gRNAs and (2) intron 12 plus 3’ gRNAs ( Figure S1A and S1B ). To generate the compound heterozygous lines for POGZ locus, hiPSCs with POGZ whole gene deletion (del1) were transfected with intron 12 plus 3’ gRNAs ( Figure S1A and S1B ). gRNAs and target editions for CRISPR engineering of SNC2A and MEF2C gene deletions were previously described 33 , 51 . 48 h after transfection, cells were separated by fluorescence-activated cell sorting (FACS) and during the first four days after FACS, hiPSCs are cultured in Essential 8 medium supplemented with CloneR (StemCell Technologies), as previously described 52 . In the CRISPR edition experiment to generate POGZ compound heterozygote lines, a wild type hiPSC line was transfected with Cas9 and GFP, subjected to FACS, and the resultant clones were used as wild type controls in the subsequent experiments. Once individual colonies were available, they were split into two populations, one for DNA extraction and genotyping (see section below) and one for future propagation. Successfully edited, clonal colonies were expanded as well as CRISPR-exposed unedited clones from the same transfection and sorting experiment, which were used as controls in downstream analysis. All replicates of edited and control clones were independently submitted to TRA-1-60 sorting, using anti-TRA-1-60 microbeads (Miltenyi Biotec) to select pluripotent cells and only selected cells were used in subsequent differentiation procedures. Characterization of CRISPR-mediated edits Prior to clone expansion, genomic DNA from individual hiPSC clones was isolated using Quick-DNA 96 kit (Zymo Research). Each clone was genotyped using PCR followed by Sanger sequencing to identify colonies that acquired desired deletions ( Supplementary Figure 1A and 1B , primer sequences are described in Supplementary Table 11 ). 16 out of 22 deletions’ junction points presented with insertions and indels involving few base pairs, ranging from 1 to 26Bp, none of them affecting coding regions or splice sites. To ensure clonality and determine gene dosage, each clone was genotyped by digital droplet PCR (Bio-Rad) using POGZ TaqMan copy number probe (Thermo Fisher Scientific, assay Hs00213580_cn), and Rnase P TaqMan Copy Number Reference Assay (Thermo Fisher Scientific) for normalization. This clone genotyping information and total clone survival rate after FACS was used to calculate the editing efficiency for each condition ( Supplementary Table 12 ). All successfully edited clones, and an equal number of control clones (i.e. unedited CRISPR-exposed clones from the same FACS batch) were expanded for future functional analysis. During clone expansion, four passages after FACS, genomic DNA from each edited and control clone was re-extracted using DNeasy Blood & Tissue Kit (Qiagen) and copy number state was re-evaluated using the same ddPCR assays ( Supplementary Figure 1C ). At this passage, the CRISPR cut sites in wild type alleles from heterozygous edited clones and control clones were assessed by PCR and sanger sequencing (primer sequences are described in Supplementary Table 11 ). Two clones with heterozygous deletions presented with a 6Bp indel and a 1Bp insertion at intron 12 and 3’ to POGZ cut sites, respectively, and none of these events affected coding regions or splice sites. All control clones used in subsequent experiments did not show insertions or indels at Cas9 cut sites. Genotyping of SNC2A and MEF2C gene deletions were previously described 33 , 51 . Replicates and batches design for neuronal differentiation experiments with POGZ models Differentiation of POGZ heterozygous hiPSC lines into neuronal lines was conducted on batches of six CRISPR/Cas9 edited replicates and six control replicates per POGZ deletion per background line, resulting in four batches of 12 clones in total ( Supplementary Table 13 ). Differentiation of POGZ compound heterozygous hiPSC lines into neuronal lines was conducted on one batch of six compound heterozygous replicates, six heterozygous whole gene deletion replicates and six wild type control replicates ( Supplementary Table 13 ). Independently edited clones were used as replicates in the following experiments when possible. For wild type hiPSCs CRISPR transfection/FACS experiments that generated less than six clones with heterozygous deletions, one or two edited clones were split during expansion to reach the total of six replicates ( Supplementary Table 13 ). Since CRISPR transfection/ FACS experiments to generate compound heterozygote lines generated a single edited clone, this one clone was split into six replicates for future experiments ( Supplementary Table 13 ). After neuronal differentiation, given a certain cell type, the replicates for RNA-seq and ATAC-seq experiments were harvested in parallel. Neural induction of hiPSC lines and expansion of NSC Neural induction of hiPSC lines was performed with PSC Neural Induction Medium (Thermo Fisher Scientific) according to the manufacturer’s instructions. hiPSC clones were at 5-8 passages after CRISPR transfection when neural induction was initiated. Briefly, on day 0 of neural induction, ∼24 h after the cells were split, the Essential 8 medium was replaced with PSC neural induction medium, which contains neurobasal medium and PSC neural induction supplement. The neural induction medium was exchanged every other day from day 0 to day 5 of neural induction. On day 7, NSC (P0) were harvested and expanded in neural expansion medium containing 50% neurobasal medium (Thermo Fisher Scientific), 50% advanced DMEM/F12 (Thermo Fisher Scientific), and neural induction supplement (Thermo Fisher Scientific) on Matrigel. At passage 4, all clones were submitted to NCAM sorting using anti-PSA-NCAM microbeads (Miltenyi Biotec) to select neural cells and only the selected cells were plated at passage 5. At passage 7, NSC were plated on coverslips and analyzed for Nestin, PAX6, SOX1 and SOX2 expression by immunostaining using Human Neural Stem Cell Immunocytochemistry Kit (Thermo Fisher Scientific) according to manufacturer’s instructions ( Supplementary Figure 16 ). Expanded NSC at passage 7 were harvested and used for subsequent assays. Validation of POGZ expression reduction After neural differentiation, the impact of the edits on transcript and protein abundance was measured by digital droplet PCR (Bio-Rad) and western blot, respectively. For digital droplet PCR gene expression assays, RNA was isolated with Trizol reagent following manufacturer’s instructions, and cDNA was synthesized using Maxima H Minus cDNA Synthesis kit (Thermo Fisher Scientific). POGZ transcript abundance was accessed with TaqMan gene expression probes (Thermo Fisher Scientific, assays Hs00374850_ m1 and Hs00418559_m1) and normalized by GUSB gene expression (Hs00939627_m1). For western blot assays, fractionated protein extraction was performed using NE PER Nuclear and Cytoplasmic Extraction Kit (Thermo Fisher Scientific). Nuclear extracts were resolved on NuPAGE™ 3-8% Tris-Acetate Gels (Thermo Fisher Scientific) and wet-transferred to PVDF membrane (Millipore Sigma). Membranes were blocked with Odyssey Blocking Buffer (LI-COR Bioscience) for 60 minutes at room temperature and then probed overnight at 4°C with anti-POGZ (Abcam ab171934, 1:300) and anti-PCNA (Abcam ab29, 1:1000), which was used for normalization. After three washes, membranes were incubated with IRDye secondary antibodies (LI-COR Bioscience) ( Supplementary Figure 1D ). Induction of Ngn2-neurons from hiPSC lines Ngn2-neuronal induction was performed as previously described 53 with modifications. Briefly, hiPSCs were seeded at low density and transduced with a lentivirus expressing TetO-Ngn2-GFP-Puro or TetO-Ngn2-Puro along with rtTA. Twenty-four hours after transduction, doxycycline was added to initiate Ngn2 expression, and then 24 h later the cells were selected with puromycin. hiPSC-derived Ngn2 -induced neurons were cultured in neuronal maintenance medium supplemented with BDNF and NT-3 growth factors. Subsequent experiments were performed with 3-week-old Ngn2 -induced neurons. RNA-seq on neuronal models RNA was isolated with Trizol reagent (Thermo Fisher Scientific) and Phasemaker Tubes (Invitrogen) following manufacturer’s instructions. Both POGZ and SCN2A RNA-seq libraries were prepared using 200 ng of total RNA using a TruSeq Stranded mRNA Sample Prep Kit (Illumina cat # RS-122-2102). Libraries were multiplexed, pooled and sequenced on an Illumina NovaSeq, generating an average of 40.7 million paired-end 150-cycle reads for 133 POGZ samples and an average of 33.5 million paired-end 150-cycle reads for 38 SCN2A samples. RNA-seq data analysis Quality of sequence reads was assessed by fastQC (version 0.10.1) ( https://www.bioinformatics.babraham.ac.uk/projects/fastqc/ ). Raw sequence reads were trimmed against Illumina adapters using Trimmomatic (v. 0.36) with parameters ILLUMINACLIP:adapter. fa:2:30:10 LEADING:3 TRAILING:3 SLIDINGWINDOW:4:15 MINLEN:75 ( https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4103590/ ). We generated gene-based counts for RNA-seq libraries by aligning sequence reads to the human reference genome, GRCh38 (v92), and relying on Ensembl gene annotations of this reference genome by using STAR (version 2.5.2A) 54 with parameters ‘‘–outSAMunmapped Within –outFilterMultimapNmax 1 –outFilterMismatchNoverLmax 0.1 –alignIntronMin 21 – alignIntronMax 0 –alignEndsType Local –quantMode GeneCounts –twopassMode Basic’’. We assessed the quality of alignments by using custom scripts in Picard Tools, RNA-seqC 55 , and SamTools 56 . These quality-checking assessments and exploratory analyses identified two outlier samples from POGZ samples: one NSC del2 sample from MGH2069 background (NSC_03_2020_KOK25_14) and one iN WT sample from GM08830 background (iN_05_2021_ 8330_WT_C6), and three outlier samples from SCN2A samples: two edited samples with SCN2A heterozygous deletion (CD1-55, CD3-50) and one WT sample (CD3-43). To validate the deletion introduced by CRISPR on POGZ and SCN2A loci, raw exon-level coverage counts was generated using DEXseq 57 on all annotated transcript isoforms ( Supplementary Tables 14 and 15 ). Raw exon-coverage counts were further normalized by exon length and number of uniquely mapped reads which were retrieved from STAR aligner’s log file for each library, following the formula 10 9 x raw_count_exon ij / (Uniquely Mapped Reads j x Exon Length i ) where index i denotes the exon number in the j th library. For each library, normalized coverage of deletion region was statistically compared to normalized coverage of exons outside the deletion region applying one-tailed t-test. In each comparison of target gene deletion vs WT samples, genes with more than 0.5 counts per million (CPM) in at least 50% of samples in either condition were analyzed. In gene-level CPM calculations, the number of uniquely mapped reads output from STAR aligner was used as library size. Differential expression analysis Differential expression analysis for POGZ iN (heterozygous deletion vs WT) and SCN2A iN (heterozygous deletion vs WT) RNA-seq libraries were performed using DESeq2 58 + SVA 59 pipeline in which surrogate variables were estimated using R/Biocondutor package, SVA with ∼genotype as the full model and ∼1 as reduced model. In differential expression analysis of POGZ iN samples, del1 and del2 genotypes were combined under del for each background, and DESeq2 + SVA pipeline was applied to each background. Shared genes that were significant at FDR<0.1 and showed the same direction of effect in both backgrounds comprised iN DEG. NSC DEG were identified similarly as shared genes that were significant at FDR<0.1 and showed the same direction of effect in both backgrounds. However, differential expression of genes in each background was assessed by integrating del1 and del2 through meta-analysis. In doing so, in each background, differential expression of each gene due to del1 and del2 was assessed by z scores. These z scores were combined under Stouffer’s weighted z score method and converted to p-value following normal distribution. Adjusted p-value (FDR) were calculated separately for each background, following Benjamini-Hochberg procedure. In each background, DEG were defined among genes that were significant at FDR<0.1 and showed the same direction of effect due to del1 and del2. Only DESeq2 was used for differential expression analyses of POGZ (compound heterozygous vs WT) in NSC and iN as principal component analysis revealed that compound heterozygous samples and control samples were distinctly clustered. Co-expression analysis Co-expression network analysis was performed using R package WGCNA 28 for each cell type separately using the signed network type for which we used log transformed CPM filtered and SVA corrected counts. Soft power was selected such that the scale-free topology fit (R2)>0.85. and the smallest module size was set to 50. Merge of the modules with similar eigengene profiles was performed (Similarity>75%). Module membership for each gene was re-evaluated based on the module membership p-value; genes with p-value>0.01 were marked as unassigned (gray module 0). Modules of interest in NSC and iN were identified based on statistical significance of module eigengene correlation with POGZ genotype. In co-expression analysis of chromatin and synaptic ASD/NDD genes, both heterozygous deletion and control iN samples for POGZ (23 samples) and SCN2A (35 samples) from GM8083 background were considered, whereas MEF2C iN samples (11 samples) were excluded as they were fewer compared to others. SVA corrected counts were generated by applying SVA package to counts that were normalized by DESeq2 under the full model ∼ gene_genotype, where gene_genotype consists of three groups of genotype; POGZ heterozygous deletion combining both del1 and del2, SCN2A heterozygous deletion and control samples combining control samples from POGZ del1 and del2 batches, and SCN2A batch and ∼1 as reduced model. One SCN2A WT sample (CD3-38) was identified as an outlier according to previously described procedures 60 , 61 Enrichment analysis We tested gene enrichments (e.g. DEG and co-expression modules) of DEG and co-expression module genes using one-tailed Fisher’s exact test for the curated lists of Gene Ontology (GO) Terms, Canonical Pathways including REACTOME and KEGG, Human Phenotype Ontology terms, and TF targets from mSigDB (version 2022.1.Hs) 62 , 63 , and synaptic-function associated GO terms from SynGO (v1.1) 64 . Additionally, we used phenotype-informed literature data including gene sets previously published with functional associations, mutational constraints or with neurological phenotype, and POGZ function. Gene sets with functional associations and mutational constraints are chromatin modifiers and synaptic genes from SynGO (v1.1) 64 , chromatin modifiers 65 , mRNA targets of fragile X messenger ribonucleoprotein (FMRP targets) 66 , and loss-of-function (LoF)-constrained genes 8 . Gene sets associated with neurological phenotype are ASD-risk genes 1 and NDD-risk genes 3 . Gene sets related to POGZ function are DEG from multiple brain tissues of Pogz -deficient mouse models including DEG from mouse hippocampus and cerebellum 21 , and DEG from mouse basal ganglia and cortex 22 , as well as mouse Pogz targets 23 and human POGZ targets 22 . The resulting p-value were corrected for multiple tests using Benjamini-Hochberg correction 67 . ATAC-seq on neuronal cell models ATAC-seq was applied to NSC and iN with the following genotypes: six compound heterozygous clones, six heterozygous whole gene deletion clones and six wild type control clones. ATAC-seq was performed as previously described 68 with minor modifications. Briefly, 50,000 cells were washed with ice-cold PBS, then re-suspended in 50ul cold Lysis Buffer [10 mM Tris-HCl (pH 7.5), 10 mM NaCl, 3 mM MgCl2, 0.1% (vol/vol) IGEPAL, 0.1% (vol/vol) Tween-20, 0.01% (vol/vol) Digitonin] and kept on ice for 3 minutes. 1 ml Wash Buffer [10 mM Tris-HCl(pH 7.5), 10 mM NaCl, 3 mM MgCl2, 0.1% (vol/vol) Tween-20] was added and samples were centrifuged for 10 minutes at 4°C and 500g. After lysis, the nuclei were re-suspended in 50 ul transposition reaction mix using reagents from Tagment DNA TDE1 Enzyme and Buffer Kits (Illumina) [2X Tagment DNA Buffer, 0.1% (vol/vol) Tween-20, 0.01% (vol/vol) Digitonin, 2.5 μl Tn5 Transposase] and incubated for 30 min at 37°C with constant agitation at 1000rpm. The transposed DNA was purified with MinElute Reaction Cleanup kit (Qiagen) and eluted in 20ul Elution Buffer. DNA fragments were PCR pre-amplified using NEBNext High-Fidelity 2X PCR Master Mix (New England Biolabs) for 5 cycles initially, and one tenth of the volume was removed for qPCR amplification for 20 cycles. A plot of R vs cycle number was generated, and the number of cycles required to reach one third of the maximum R was determined for each sample. The pre-amplified ATAC-seq libraries were then amplified for the calculated additional cycles. Agencourt AMPure XP beads (Beckman Coulter) were used for size selection according to manufacturer’s instructions with 1.8:1 beads to DNA ratio. The size-selected libraries were run on the TapeStation D1000 tape and reagents (Agilent) in order to determine fragment size profile. Libraries were multiplexed, pooled and sequenced on an Illumina NovaSeq 6000, generating an average of 50 million paired-end 50-cycle reads for each sample. ATAC-seq analysis Data processing ATAC-seq data were first processed based on their sequencing batches. For each batch, the processing followed standard ENCODE ATAC-seq workflow ( https://docs.google.com/document/d/1f0Cm4vRyDQDu0bMehHD7P7KOMxTOP-HiNoIvL1VcBt8/edit ). Briefly, for each library within a batch, the adapters were trimmed, and the reads were aligned to the GRCh38 human genome assembly including mitochondria using Bowtie2 (v2.3.1). Post-alignment quality control approaches were applied, including investigating mitochondria fraction, GC bias, fragment length, phantom peaks and removing PCR duplicates using customized scripts and PicardTools (v1.128). The BAM file was further converted to TagAlign format, where the coordinates adjustment due to Tn5 cleaves was applied by shifting the start positions four bases toward the downstream (i.e. +4 for reads aligned to the positive strand and -5 for reads aligned to the negative strand in a TagAlign file). This TagAlign file was then shuffled and randomly splitted into two files with equal number of lines, which made the two pseudo replicates of that library (i.e. pseudo replicate 1 and 2). For libraries with the same genotype in a batch, they were considered as biological replicates. For each genotype, all the biological replicates, all the pseudo replicate 1’s and all the pseudo replicate 2’s were pooled respectively, generating three extra artificial libraries. Peak calling was then carried out within each genotype on all the libraries, their pseudo replicates and the three pooled artificial libraries, using MACS2 69 version 2.2.7.1 with “-g hs -p 0.01 --nomodel --shift -75 --extsize 150 --keep-dup all -B –SPMR --call-summits” options. This adds up to 3*(N+1) calls for a genotype with N libraries sequenced. Only the top 300,000 peaks with the smallest p-value from each call were retained for the downstream analyses. To find consensus peaks in each genotype across all the biological replicates, pseudo replicates and artificially pooled libraries, a naive overlap approach was applied, using an overlap cutoff of 50%. This approach searched for overlapped peaks between any two biological replicates that further overlapped with the peaks from the artificially pooled library (i.e. the artificial library where all the biological replicates were pooled); and (b) the overlapped peaks between the two pooled pseudo replicates that further overlapped with the peaks from the artificially pooled library (i.e. the artificial library where all the biological replicates were pooled). This generates 0.5*N*(N-1)+1 overlap sets for a genotype with N libraries sequenced. Each overlap set also excluded the genomic regions recorded in the ENCODE blacklist (hgdownload. cse.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeMapability/). For each genotype within a sequencing batch, the overlap set with the greatest number of surviving peaks in (a) was considered as the set of conservative consensus peaks while the overlap set with the greatest number of surviving peaks in both (a) and (b) was considered as the set of optimal consensus peaks. Note, only the conservative consensus peaks were used throughout this study. Finally, the consensus peaks were merged across sequencing batches and genotypes to generate a merged peak set per cell type per iPSC background. Differential Accessibility The bioconductor package DiffBind (v3.4.11) 70 was applied to identify DARs between genotypes within each hiPSC background, using the final alignments of ATAC-seq (i.e. BAM files) and the consensus peaks per genotype per sequencing batch as the input. POGZ compound heterozygous peaks were compared to peaks in wild type clones, using dba.contrast function, model design=“∼Treatment” and contract=c(“Treatment”, “DEL”, “WT”). “Treatment” represents sample genotype (del1, del1_del2, or WT). POGZ heterozygous deletion peaks in GM08830 and MGH2069 were compared to their corresponding peaks in wild type clones, respectively, using model design=∼Factor+Treatment and contrast=c(“Treatment”, “DEL”, “WT”). “Factor” represents the sample genotype (del1, del2 or WT), “Treatment” represents combined genotype, either DEL or WT. Regions with an FDR<0.1 were considered significantly DARs (see Supplementary Information for more details). DARs from different backgrounds in the same cell type were then merged by taking their overlaps to make the final cell-type-specific DARs. DARs target genes were identified as gene promoters (2000 bp upstream to 500 bp downstream of the transcription start site) that overlapped with cell-type-specific DARs. One gene promoter might overlap with more than one DARs and the DARs contained in the same promoter region might be regulated in opposite directions. Transcription factor footprint TOBIAS (v0.12.10) 71 was used to call TF footprint from ATAC-seq. First, the final alignments (i.e. BAM files) for samples from the same genotype within the same sequencing batch were merged. The correction for Tn5 shifting was achieved by calling the “ATACorrect” function of TOBIAS on each merged BAM file, within the corresponding merged peak set (i.e. per cell type per iPSC background) generated in the previous step (see “Transcription factor footprint” section in Methods). This generated a corrected BIGWIG file for ATAC-seq coverage, which was further scored by calling the “FootprintScores” function of TOBIAS per genotype per sequencing batch. The scored BIGWIG files were merged across sequencing batches using the USCS script “bigWigMerge” to generate footprint profiles per genotype per iPSC background. Then the “BINDetect” function was used to predict TF binding using the sequence motifs recorded in the JASPER database and compare the differential bindings between genotypes. For each TF’s motif, this generated two levels of changes between genotypes, namely: 1) locus level, which compares footprint difference between genotypes for each specific genomic locus that matches the TF motif; and 2) genome-wide level, which summarizes the global footprint difference between genotypes across all the loci of that TF. A differentially bound transcription factor (DTF) was defined as the genome-wide summarized changes of any motif in a single TF either 1) ranked top 5% of the largest -log10(pvalue); or 2) top and bottom 5% of differential binding score between genotypes. Finally, overlapped DTFs from the GM08830 and MGH2069 backgrounds were used as the final DTFs for POGZ heterozygous. Direct and Indirect POGZ regulation To generate the list of genes directly regulated by POGZ, POGZ-occupied loci in the human fetal cortex 22 were used. Genes whose promoter overlapped with the POGZ-occupied loci were annotated as the direct POGZ targets in the human cortex. To generate the list of genes indirectly regulated by POGZ, each individual binding site belonging to a DTF was compared between genotypes. Those binding sites with substantial changes (predicted bound in at least one condition of DEL and WT, and the log2 fold change ranked within top or bottom 10%) of each DTF were retained. Any of these retained sites further overlapping with a gene promoter was considered as an indirect POGZ target. Multi-Electrode Array (MEA) analysis 48-well MEA plates (Axion Biosystems, M768-tMEA-48B) were prepared by coating wells with 50 µL 0.1% Polyethylenimine (PEI) (Millipore-Sigma, 408727) dissolved in Borate Buffer (Boston Bioproducts, BB-66-500) and sterile filtered with a 0.2μm filter prior to use. Coated wells were incubated at 37ºC for 1hr. Wells were then washed 6-10x with tissue culture-grade water (Boston Bioproducts, 7732-18-5) to remove residual PEI solution. Once excess water was removed, plates were stored at RT overnight to dry out completely. To prepare iN for electrophysiology readings we largely followed manufacturer protocol with a few modifications. Briefly, TRA1+ hiPSC lines underwent 1 post-thaw passage prior to further manipulation. All hiPSC lines for MEA underwent Ngn2 transduction as a single batch to eliminate viral transduction batch effect using the same transduction and selection steps described in “ Induction of Ngn2 -neurons from hiPSC lines ” above. On Day 4 of iN differentiation, cells were detached using Accutase (STEMCELL, 07920), counted using a Countess Cell Counter (Thermo Fisher Scientific, A27977), and resuspended in NMM supplemented with BDNF, NT-3, Doxycycline and Laminin as outlined in manufacturer protocols (Axion Biosystems: Culturing Human hiPSC-derived Excitatory Neurons on Microelectrode Arrays: Maestro Pro MEA) in order to distribute 6x10 4 cells per well in a volume of 10 µL directly to the center of each MEA plate well. Plated cells were left to adhere for 1hr at 37ºC prior to the addition of 200 µL NMM supplemented with BDNF, NT-3, and Doxycycline per well. 200 µL NMM supplemented with BDNF, NT-3, Doxycycline, and Ara-C was added on Day 5 for a total volume of 400 µL per well. Each well underwent a half media change (removal and addition of 200 µL) every 3 days with NMM supplemented with BDNF and NT-3 until Day 24. Doxycycline use was discontinued after Day 10 and Ara-C was only added on Day 6. Day 27 – Day 51 cells continued to undergo half media change every 3 days, but media was changed to BrainPhys (following manufacturer protocol: Axion Biosystems: Culturing Human hiPSC-derived Excitatory Neurons on Microelectrode Arrays: Maestro Pro MEA) to promote spiking activity. MEA plate readings were conducted every 3 days, 22-24 hours following each media change. Readings were conducted at 37ºC for 15min using Axion Biosystems Maestro Pro. Manufacturer set thresholds for spike and network burst calling were used. For data interpretation, only wells with at least 4 active electrodes were filtered. A total of 9 readings were performed from Day 24 to Day 51 and the reading with the highest number of clones with at least 4 active electrodes across all wells and genotypes was selected for statistical analysis ( Supplementary Figure 17 ). Weighted mean firing rate was compared between genotypes using the Wilcoxon test. Incucyte analysis The inner 60 wells of 96-well plates were prepared by coating wells with PLO/Laminin and iN were prepared for Incucyte recordings following the protocol described in “ Ngn2-induced neurons from hiPSC lines ” above, with volume adjustments. All hiPSC lines for Incucyte underwent Ngn2 transduction at the same batch performed for MEA readings to eliminate viral transduction batch effect. On Day 4 of iN differentiation, cells were detached using Accutase (STEMCELL, 07920), counted using a Countess Cell Counter (ThermoFisher, A27977), and resuspended in NMM supplemented with BDNF, NT-3 and Doxycycline in order to distribute 12x10 3 cells in a volume of 75 µL per well. 75 µL NMM supplemented with BDNF, NT-3, Doxycycline, and Ara-C was added on Day 5 for a total volume of 400 µL per well. Each well underwent a half media change (removal and addition of 75 µL) every 3 days with NMM supplemented with BDNF and NT-3 until Day 27. Doxycycline use was discontinued after Day 10 and Ara-C was only added on Day 6. Stain-free Incucyte recordings were performed using a IncuCyte S3 with images acquisition every four hours, in three timepoints: Days 11-13, Days 18-20, and Days 25-27. Neurotrack analysis was performed in IncuCyte software (version 2020B) largely following default parameters, with a few modifications to increase sensitivity in detecting iN neurites and omit cellular debris. Neurotrack settings were “minimal cell width”=11, “minimal cell body cluster”=125, and “neurite sensitivity”=0.7. Neurite length and neurite branch points metrics were normalized by the area of cell body clusters in each well to eliminate variability due to artifactual cell counting differences. A total of 9 readings were performed from Day 11 to Day 27 and the reading with the highest overall neurite length across all wells and genotypes was selected for statistical analysis. Neurite length and neurite branch points were compared between genotypes using the Wilcoxon test. DECLARATIONS Authors’ contribution Study design: M.M.O., Y.L., S.E., D.G., D.R., K.B., J.F.G., M.E.T. Experiments: M.M.O., Y.L., S.E., D.G., R.B., K.M., K.O.K, P.M.B., G.X., C.L., A.L., R.Y., M.S., D.L., B.C., C.E.F.E., D.J.C.T., D.R., K.B., J.F.G., M.E.T. Data interpretation: M.M.O., Y.L., S.E., D.G., D.R., K.B., J.F.G., M.E.T. Manuscript writing: M.M.O., Y.L., S.E., D.G., J.F.G., M.E.T. All authors have reviewed and accepted final version of the manuscript. Declaration of interests M.E.T. receives research funding and/or reagents from Levo Therapeutics, Microsoft Inc, and Illumina Inc, is a paid consultant for BridgeBio, and is co-founder of First Genomic Insights, LLC. J.F.G. is a paid consultant to Harness Therapeutics and has consulted for Biogen, Inc., Pfizer, Inc., and Wave Biosciences, Inc. in the past five years. SUPPLEMENTAL MATERIAL Supplemental Information POGZ downregulation from RNA-seq data PCA and SV correction description in each cell type, justifying strategies for data integration in iN and in NSC WGCNA for POGZ models – modules description Differential accessibility regions (DARs) for POGZ cellular models TF expressed in NSC and iN and DTFs reproducibility across backgrounds Detection of DTF regulatory targets and their reproducibility across backgrounds DEG comparisons between POGZ, MEF2C and SCN2A heterozygous neuronal models Supplemental Figures Supplementary Figures 1-25 Supplemental Tables Supplementary Tables 1-15 Acknowledgements This research was supported by grants from the National Institutes of Health (NIH, R01MH123155, 5R01MH115957, R01HD096326, RM1MH132648, U01HG011755), the Simons Foundation Autism Research Initiative (SFARI, #573206 and #1009802). M.M.O. was supported by Autism Speaks Postdoctoral Fellowship (11815). R.Y. was supported by the Mass General Hospital Fund for Medical Discovery. P.M.B. was supported by the NIH (K08NS117891). D.G. was supported by the NIH (K99NS118109, R00NS118109-03). Funder Information Declared NIH Common Fund , R01MH123155 , 5R01MH115957 , R01HD096326 , RM1MH132648 , U01HG011755 , K08NS117891 Simons Foundation, https://ror.org/01cmst727 , 573206 , 1009802 Autism Speaks, https://ror.org/04bkad313 , 11815 REFERENCES 1. ↵ Fu , J. M. et al. Rare coding variation provides insight into the genetic architecture and phenotypic context of autism . Nat Genet ( 2022 ) doi: 10.1038/S41588-022-01104-0 . OpenUrl CrossRef 2. Grove , J. et al. Identification of common genetic risk variants for autism spectrum disorder . Nat Genet 51 , 431 – 444 ( 2019 ). OpenUrl CrossRef PubMed 3. ↵ Kaplanis , J. et al. Evidence for 28 genetic disorders discovered by combining healthcare and research data . Nature 586 , 757 – 762 ( 2020 ). OpenUrl CrossRef PubMed 4. ↵ Lowther , C. et al. Systematic evaluation of genome sequencing for the diagnostic assessment of autism spectrum disorder and fetal structural anomalies . Am J Hum Genet 110 , 1454 – 1469 ( 2023 ). OpenUrl CrossRef PubMed 5. ↵ Feliciano , P. et al. Exome sequencing of 457 autism families recruited online provides evidence for autism risk genes . NPJ Genom Med 4 , 19 ( 2019 ). OpenUrl PubMed 6. De Rubeis , S. et al. Synaptic, transcriptional and chromatin genes disrupted in autism . Nature ( 2014 ) doi: 10.1038/nature13772 . OpenUrl CrossRef PubMed Web of Science 7. ↵ Satterstrom , F. K. et al. Large-Scale Exome Sequencing Study Implicates Both Developmental and Functional Changes in the Neurobiology of Autism . Cell 180 , 568 - 584 .e23 ( 2020 ). OpenUrl CrossRef PubMed 8. ↵ Karczewski , K. J. et al. The mutational constraint spectrum quantified from variation in 141,456 humans . Nature 581 , 434 – 443 ( 2020 ). OpenUrl CrossRef PubMed 9. ↵ Moyses-Oliveira , M. , Yadav , R. , Erdin , S. & Talkowski , M. E. New gene discoveries highlight functional convergence in autism and related neurodevelopmental disorders . Curr Opin Genet Dev 65 , 195 – 206 ( 2020 ). OpenUrl CrossRef PubMed 10. Willsey , H. R. , Willsey , A. J. , Wang , B. & State , M. W. Genomics, convergent neuroscience and progress in understanding autism spectrum disorder . Nat Rev Neurosci 23 , 323 – 341 ( 2022 ). OpenUrl CrossRef PubMed 11. ↵ Willsey , H. R. et al. Parallel in vivo analysis of large-effect autism genes implicates cortical neurogenesis and estrogen in risk and resilience . Neuron 109 , 788 - 804 .e8 ( 2021 ). OpenUrl CrossRef PubMed 12. Brennand , K. J. & Talkowski , M. E. Xenopus models suggest convergence of gene signatures on neurogenesis in autism . Neuron 109 , 743 – 745 ( 2021 ). OpenUrl PubMed 13. ↵ O’Roak , B. J. et al. Recurrent de novo mutations implicate novel genes underlying simplex autism risk . Nat Commun ( 2014 ) doi: 10.1038/ncomms6595 . OpenUrl CrossRef PubMed 14. Talkowski , M. E. et al. Sequencing chromosomal abnormalities reveals neurodevelopmental loci that confer risk across diagnostic boundaries . Cell ( 2012 ) doi: 10.1016/j.cell.2012.03.028 . OpenUrl CrossRef PubMed 15. Neale , B. M. et al. Patterns and rates of exonic de novo mutations in autism spectrum disorders . Nature ( 2012 ) doi: 10.1038/nature11011 . OpenUrl CrossRef PubMed Web of Science 16. Sanders , S. J. et al. De novo mutations revealed by whole-exome sequencing are strongly associated with autism . Nature ( 2012 ) doi: 10.1038/nature10945 . OpenUrl CrossRef PubMed Web of Science 17. ↵ Wright , C. F. et al. Genetic diagnosis of developmental disorders in the DDD study: a scalable analysis of genome-wide research data . Lancet 385 , 1305 – 1314 ( 2015 ). OpenUrl CrossRef PubMed 18. ↵ Stessman , H. A. F. et al. Disruption of POGZ Is Associated with Intellectual Disability and Autism Spectrum Disorders . Am J Hum Genet 98 , 541 – 552 ( 2016 ). OpenUrl CrossRef PubMed 19. ↵ White , J. et al. POGZ truncating alleles cause syndromic intellectual disability . Genome Med 8 , ( 2016 ). 20. ↵ Matsumura , K. et al. Pathogenic POGZ mutation causes impaired cortical development and reversible autism-like phenotypes . Nat. Commun . 11 , 859 ( 2020 ). OpenUrl CrossRef PubMed 21. ↵ Suliman-Lavie , R. et al. Pogz deficiency leads to transcription dysregulation and impaired cerebellar activity underlying autism-like behavior in mice . Nat Commun 11 , ( 2020 ). 22. ↵ Markenscoff-Papadimitriou , E. et al. Autism risk gene POGZ promotes chromatin accessibility and expression of clustered synaptic genes . Cell Rep 37 , ( 2021 ). 23. ↵ Sun , X. , Cheng , L. & Sun , Y. Autism-associated protein POGZ controls ESCs and ESC neural induction by association with esBAF . Mol Autism 13 , ( 2022 ). 24. ↵ Agirre , E. et al. A chromatin code for alternative splicing involving a putative association between {CTCF} and {HP1$α$} proteins . BMC Biol . 13 , 31 ( 2015 ). OpenUrl CrossRef PubMed 25. ↵ Ostapcuk , V. et al. Activity-dependent neuroprotective protein recruits {HP1} and {CHD4} to control lineage-specifying genes . Nature 557 , 739 – 743 ( 2018 ). OpenUrl CrossRef PubMed 26. ↵ Gudmundsdottir , B. et al. POGZ Is Required for Silencing Mouse Embryonic β-like Hemoglobin and Human Fetal Hemoglobin Expression . Cell Rep 23 , 3236 – 3248 ( 2018 ). OpenUrl CrossRef PubMed 27. ↵ Nozawa , R. S. et al. Human POGZ modulates dissociation of HP1α from mitotic chromosome arms through Aurora B activation . Nat Cell Biol 12 , 719 – 727 ( 2010 ). OpenUrl CrossRef PubMed Web of Science 28. ↵ Langfelder , P. & Horvath , S. WGCNA: An R package for weighted correlation network analysis . BMC Bioinformatics 9 , ( 2008 ). 29. ↵ Valencia , A. M. et al. Landscape of mSWI/SNF chromatin remodeling complex perturbations in neurodevelopmental disorders . Nat Genet 55 , 1400 – 1412 ( 2023 ). OpenUrl CrossRef PubMed 30. ↵ Buenrostro , J. D. , Giresi , P. G. , Zaba , L. C. , Chang , H. Y. & Greenleaf , W. J. Transposition of native chromatin for fast and sensitive epigenomic profiling of open chromatin, DNA-binding proteins and nucleosome position . Nat Methods ( 2013 ) doi: 10.1038/nmeth.2688 . OpenUrl CrossRef PubMed Web of Science 31. ↵ Li , F. , Tian , J. , Zhang , L. , He , H. & Song , D. A multi-omics approach to reveal critical mechanisms of activator protein 1 (AP-1) . Biomedicine and Pharmacotherapy vol. 178 Preprint at doi: 10.1016/j.biopha.2024.117225 ( 2024 ). OpenUrl CrossRef PubMed 32. ↵ Joo , J. Y. , Schaukowitch , K. , Farbiak , L. , Kilaru , G. & Kim , T. K. Stimulus-specific combinatorial functionality of neuronal c-fos enhancers . Nat Neurosci 19 , 75 – 83 ( 2016 ). OpenUrl CrossRef PubMed 33. ↵ Mohajeri , K. et al. Transcriptional and functional consequences of alterations to MEF2C and its topological organization in neuronal models . Am J Hum Genet 109 , 2049 – 2067 ( 2022 ). OpenUrl CrossRef PubMed 34. ↵ Deneault , E. et al. Complete Disruption of Autism-Susceptibility Genes by Gene Editing Predominantly Reduces Functional Connectivity of Isogenic Human Neurons . Stem Cell Reports 11 , 1211 – 1225 ( 2018 ). OpenUrl PubMed 35. Marie , C. et al. Oligodendrocyte precursor survival and differentiation requires chromatin remodeling by Chd7 and Chd8 . Proc. Natl. Acad. Sci. U. S. A . 115 , E8246 – E8255 ( 2018 ). OpenUrl Abstract / FREE Full Text 36. ↵ Sessa , A. et al. SETD5 Regulates Chromatin Methylation State and Preserves Global Transcriptional Fidelity during Brain Development and Neuronal Wiring . Neuron 104 , 271 - 289 .e13 ( 2019 ). OpenUrl CrossRef PubMed 37. ↵ Courchesne , E. , Gazestani , V. H. & Lewis , N. E. Prenatal Origins of ASD: The When, What, and How of ASD Development . ( 2020 ) doi: 10.1016/j.tins.2020.03.005 . OpenUrl CrossRef 38. ↵ Miller , J. A. et al. Transcriptional landscape of the prenatal human brain . Nature ( 2014 ) doi: 10.1038/nature13185 . OpenUrl CrossRef PubMed Web of Science 39. ↵ Li , M. et al. Integrative functional genomic analysis of human brain development and neuropsychiatric risks . Science (1979) 362 , ( 2018 ). 40. ↵ Finucane , H. K. et al. Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types . Nat Genet 50 , 621 – 629 ( 2018 ). OpenUrl CrossRef PubMed 41. ↵ Velmeshev , D. et al. Single-cell genomics identifies cell type–specific molecular changes in autism . Science (1979) 364 , 685 – 689 ( 2019 ). OpenUrl Abstract / FREE Full Text 42. ↵ Yap , E.-L. & Greenberg , M. E. Activity-Regulated Transcription: Bridging the Gap between Neural Activity and Behavior . Neuron 100 , 330 – 348 ( 2018 ). OpenUrl CrossRef PubMed 43. ↵ Wenderski , W. et al. Loss of the neural-specific BAF subunit ACTL6B relieves repression of early response genes and causes recessive autism . Proc Natl Acad Sci U S A 117 , 10055 – 10066 ( 2020 ). OpenUrl Abstract / FREE Full Text 44. ↵ Hori , K. et al. AUTS2 Regulation of Synapses for Proper Synaptic Inputs and Social Communication . iScience 23 , ( 2020 ). 45. ↵ Trakoshis , S. et al. Intrinsic excitation-inhibition imbalance affects medial prefrontal cortex differently in autistic men versus women . Elife 9 , 1 – 31 ( 2020 ). OpenUrl CrossRef PubMed 46. Antoine , M. W. , Langberg , T. , Schnepel , P. & Feldman , D. E. Increased Excitation-Inhibition Ratio Stabilizes Synapse and Circuit Excitability in Four Autism Mouse Models . Neuron 101 , 648 - 661 .e4 ( 2019 ). OpenUrl CrossRef PubMed 47. ↵ Bartley , A. F. et al. Interneuron Transcriptional Dysregulation Causes Frequency-Dependent Alterations in the Balance of Inhibition and Excitation in Hippocampus . J Neurosci 35 , 15276 – 15290 ( 2015 ). OpenUrl Abstract / FREE Full Text 48. ↵ Sheridan , S. D. et al. Epigenetic characterization of the FMR1 gene and aberrant neurodevelopment in human induced pluripotent stem cell models of fragile X syndrome . PLoS One 6 , ( 2011 ). 49. ↵ Seo , J. et al. Inhibition of p25/Cdk5 Attenuates Tauopathy in Mouse and iPSC Models of Frontotemporal Dementia . Journal of Neuroscience 37 , 9917 – 9924 ( 2017 ). OpenUrl Abstract / FREE Full Text 50. ↵ Seabra , C. M. et al. Transcriptional consequences of MBD5 disruption in mouse brain and CRISPR-derived neurons . doi: 10.1186/s13229-020-00354-1 . OpenUrl CrossRef PubMed 51. ↵ Liao , C. et al. Convergent coexpression of autism-associated genes suggests some novel risk genes may not be detectable in large-scale genetic studies . Cell genomics 3 , ( 2023 ). 52. ↵ Tai , D. J. C. et al. Engineering microdeletions and microduplications by targeting segmental duplications with CRISPR . Nat Neurosci 19 , 517 – 522 ( 2016 ). OpenUrl CrossRef PubMed 53. ↵ Ho , S.-M. , Topol , A. & Brennand , K. J. From directed differentiation’’ to ‘‘neuronal induction’’: modeling neuropsychiatric disease . Biomark. Insights 10 , 31 – 41 ( 2015 ). OpenUrl PubMed 54. ↵ Dobin , A. et al. STAR: Ultrafast universal RNA-seq aligner . Bioinformatics ( 2013 ) doi: 10.1093/bioinformatics/bts635 . OpenUrl CrossRef PubMed Web of Science 55. ↵ Deluca , D. S. et al. RNA-SeQC: RNA-seq metrics for quality control and process optimization . Bioinformatics 28 , 1530 – 1532 ( 2012 ). OpenUrl CrossRef PubMed Web of Science 56. ↵ Li , H. et al. The Sequence Alignment/Map format and SAMtools . Bioinformatics 25 , 2078 – 2079 ( 2009 ). OpenUrl CrossRef PubMed Web of Science 57. ↵ Anders , S. , Reyes , A. & Huber , W. Detecting differential usage of exons from RNA-seq data . Genome Res 22 , 2008 – 2017 ( 2012 ). OpenUrl Abstract / FREE Full Text 58. ↵ Love , M. I. , Huber , W. , & Anders , S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2 . Genome biology 15 ( 12 ), 550 ( 2014 ). OpenUrl CrossRef PubMed 59. ↵ Leek , J. T. , Johnson , W. E. , Parker , H. S. , Jaffe , A. E. & Storey , J. D. The SVA package for removing batch effects and other unwanted variation in high-throughput experiments . Bioinformatics 28 , 882 – 883 ( 2012 ). OpenUrl CrossRef PubMed Web of Science 60. ↵ Oldham , M. C. et al. Functional organization of the transcriptome in human brain . Nat Neurosci 11 , 1271 – 1282 ( 2008 ). OpenUrl CrossRef PubMed Web of Science 61. ↵ Tai , D. J. C. et al. Tissue- and cell-type-specific molecular and functional signatures of 16p11.2 reciprocal genomic disorder across mouse brain and human neuronal models . Am J Hum Genet 109 , 1789 – 1813 ( 2022 ). OpenUrl CrossRef PubMed 62. ↵ Liberzon , A. et al. Molecular signatures database (MSigDB) 3.0 . Bioinformatics 27 , 1739 – 1740 ( 2011 ). OpenUrl CrossRef PubMed Web of Science 63. ↵ Liberzon , A. et al. The Molecular Signatures Database (MSigDB) hallmark gene set collection . Cell Syst 1 , 417 – 425 ( 2015 ). OpenUrl PubMed 64. ↵ Koopmans , F. et al. SynGO: An Evidence-Based, Expert-Curated Knowledge Base for the Synapse . Neuron 103 , 217 - 234 .e4 ( 2019 ). OpenUrl CrossRef PubMed 65. ↵ Iossifov , I. et al. The contribution of de novo coding mutations to autism spectrum disorder . Nature 515 , 216 – 221 ( 2014 ). OpenUrl CrossRef PubMed Web of Science 66. ↵ Darnell , J. C. et al. FMRP stalls ribosomal translocation on mRNAs linked to synaptic function and autism . Cell 146 , 247 – 261 ( 2011 ). OpenUrl CrossRef PubMed Web of Science 67. ↵ Benjamini , Y. & Hochberg , Y. Controlling the False Discovery Rate: A Practical and Powerful Approach to Multiple Testing . Journal of the Royal Statistical Society: Series B (Methodological) 57 , 289 – 300 ( 1995 ). OpenUrl CrossRef PubMed Web of Science 68. ↵ Buenrostro , J. D. , Wu , B. , Chang , H. Y. & Greenleaf , W. J. ATAC-seq: A method for assaying chromatin accessibility genome-wide . Curr Protoc Mol Biol ( 2015 ) doi: 10.1002/0471142727.mb2129s109 . OpenUrl CrossRef PubMed 69. ↵ Zhang , Y. et al. Model-based analysis of ChIP-Seq (MACS) . Genome Biol ( 2008 ) doi: 10.1186/gb-2008-9-9-r137 . OpenUrl CrossRef PubMed 70. ↵ Ross-Innes , C. S. et al. Differential oestrogen receptor binding is associated with clinical outcome in breast cancer . Nature ( 2012 ) doi: 10.1038/nature10730 . OpenUrl CrossRef PubMed Web of Science 71. ↵ Bentsen , M. et al. ATAC-seq footprinting unravels kinetics of transcription factor binding during zygotic genome activation . Nat Commun 11 , ( 2020 ). View the discussion thread. Back to top Previous Next Posted October 27, 2025. Download PDF Supplementary Material Email Thank you for your interest in spreading the word about bioRxiv. NOTE: Your email address is requested solely to identify you as the sender of this article. Your Email * Your Name * Send To * Enter multiple addresses on separate lines or separate them with commas. You are going to email the following CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling Message Subject (Your Name) has forwarded a page to you from bioRxiv Message Body (Your Name) thought you would like to see this page from the bioRxiv website. Your Personal Message CAPTCHA This question is for testing whether or not you are a human visitor and to prevent automated spam submissions. Share CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling Mariana Moyses-Oliveira , Yating Liu , Serkan Erdin , Dadi Gao , Riya Bhavsar , Kiana Mohajeri , Kathryn O’Keefe , Philip M. Boone , Gabriela Xavier , Calwing Liao , Aiqun Li , Rachita Yadav , Monica Salani , Diane Lucente , Benjamin Currall , Celine E. F. de Esch , Derek J. C. Tai , Douglas Ruderfer , Kristen J. Brennand , James F. Gusella , Michael E. Talkowski bioRxiv 2025.10.27.684870; doi: https://doi.org/10.1101/2025.10.27.684870 Share This Article: Copy Citation Tools CRISPR-engineered deletion of POGZ alters transcription factor binding at promoters of genes involved in synaptic signaling Mariana Moyses-Oliveira , Yating Liu , Serkan Erdin , Dadi Gao , Riya Bhavsar , Kiana Mohajeri , Kathryn O’Keefe , Philip M. Boone , Gabriela Xavier , Calwing Liao , Aiqun Li , Rachita Yadav , Monica Salani , Diane Lucente , Benjamin Currall , Celine E. F. de Esch , Derek J. C. Tai , Douglas Ruderfer , Kristen J. Brennand , James F. Gusella , Michael E. Talkowski bioRxiv 2025.10.27.684870; doi: https://doi.org/10.1101/2025.10.27.684870 Citation Manager Formats BibTeX Bookends EasyBib EndNote (tagged) EndNote 8 (xml) Medlars Mendeley Papers RefWorks Tagged Ref Manager RIS Zotero Tweet Widget Facebook Like Google Plus One Subject Area Genomics Subject Areas All Articles Animal Behavior and Cognition (7619) Biochemistry (17641) Bioengineering (13865) Bioinformatics (41860) Biophysics (21408) Cancer Biology (18545) Cell Biology (25434) Clinical Trials (138) Developmental Biology (13357) Ecology (19863) Epidemiology (2067) Evolutionary Biology (24288) Genetics (15587) Genomics (22466) Immunology (17701) Microbiology (40301) Molecular Biology (17142) Neuroscience (88441) Paleontology (666) Pathology (2825) Pharmacology and Toxicology (4814) Physiology (7633) Plant Biology (15108) Scientific Communication and Education (2042) Synthetic Biology (4285) Systems Biology (9811) Zoology (2268)