The past decade has seen tremendous progress in the identification of genes associated with complex neuropsychiatric disorders, including autism spectrum disorder (ASD) and schizophrenia. Expression patterns of these genes in single cell data strongly implicate excitatory and inhibitory neurons; however, there are limited data on the brain regions involved - a critical question for neurobiology. Spatial transcriptomics provide an opportunity to perform systematic multi-regional analyses to provide insights into this question. Here, we have generated a spatial transcriptomics dataset encompassing the diverse anatomical territories of the adult mouse brain sagittal midsection. We compare neuropsychiatric gene enrichment by applying Gene Fraction Enrichment Score (GFES), a novel statistic method that controls for differing neuronal proportions across regions. ASD-associated genes identified by exome sequencing were most enriched in the thalamus followed by the cortex. Schizophrenia genes from genome-wide association studies were also enriched in the thalamus, along with the hippocampus and cortex. These findings add to the evidence that the thalamus plays a major role in neuropsychiatric disorders whilst supporting roles for the cortex and hippocampus. The results highlight shared and distinct patterns for pleiotropic brain disorders that could elucidate common underlying mechanisms and circuitry.

A single gene can encode multiple versions of a protein, dubbed isoforms, with varying functionality. Cellular control of isoform abundances is critical for multiple aspects of biology and is only partially regulated by transcript levels. While long-read sequencing facilitates transcript quantification, quantifying the resulting protein isoforms on a large scale is a major challenge, complicating biological interpretation of transcript alterations. Standard "bottom up" mass spectrometry can assess only short portions of isoforms called peptides, and these peptides often map onto more than one isoform. We introduce PAQu, a novel Bayesian method that leverages multiomic information from the peptidome and transcriptome to provide accurate estimates of isoform abundance even when peptide mapping is ambiguous. PAQu offers several advantages over existing methods in a unified framework. It provides uncertainty quantification, integrates multiomic information for improved accuracy, and provides a rigorous framework for hypothesis testing. Extensive simulations show that PAQu consistently outperforms competing methods in detecting differentially expressed protein isoforms and estimating their abundances. We use PAQu to investigate differences in isoform abundance levels between people with schizophrenia and control subjects, confirming a long held hypothesis that levels of the C4A isoform of Complement Component 4 are increased in schizophrenia while C4B is not. These results demonstrate that PAQu can identify significant variations in isoform abundance levels not previously possible.

Obsessive-compulsive disorder (OCD) is a chronic psychiatric illness associated with altered function in cortico-striatal-thalamo-cortical (CSTC) circuits. In this pilot study, we examined differential RNA expression in the thalamus using postmortem human brain tissue samples from 11 subjects with OCD and 10 unaffected subjects. We individually dissected the mediodorsal magnocellular, mediodorsal parvocellular, and ventral anterior nuclei, which participate in orbitofrontal and anterior cingulate CSTC circuits most frequently associated with OCD, and the posterior ventrolateral nucleus, which participates in premotor and motor circuits that are increasingly implicated in OCD. Preselected GABAergic, glutamatergic and ion channel genes were analyzed via qPCR. Two genes required for GABA synthesis and release, GAD1 and SLC32A1, were found to be downregulated in OCD subjects across all nuclei, and potassium channel KCNN3 was upregulated. In parallel, we performed an exploratory total RNAseq differential expression analysis. We identified few (12-52) differentially expressed genes (DEGs) in each nucleus, and only one DEG in a pooled analysis of all nuclei. No DEGs were significant after correction for multiple comparisons. Investigation by model selection indicated that OCD diagnosis was not a useful factor in modelling gene expression in our dataset. OCD was also not associated with any modules of co-expressed genes identified using weighted gene correlation network analysis. Overall, we found minimal evidence of differential RNA expression in these thalamic nuclei in OCD. These findings contrast with our previous work including many of the same subjects where we found widespread differential mRNA expression in the orbitofrontal cortex and striatum in OCD.

Autism spectrum disorder (ASD) is estimated to be up to four times as common in males as in females, yet the causes of this prevalence difference are not well established. One possible driver is genetic variation on the X chromosome, as it contains genes capable of contributing to ASD (e.g., PTCHD1, MECP2) and is known to play a role in genetic disorders with differential sex prevalence (e.g., color blindness). However, a lack of power compared to the autosomes combined with the complexities of modeling its biology have led to the X being largely overlooked in sequencing studies. Here, we develop quantitative X-linked TADA, a new model designed specifically for application to this chromosome, and use it to analyze rare variation from 50,663 individuals with ASD (and 136,670 individuals total). We find 9 genes on the X associated with ASD at a false discovery rate (FDR) < 0.05 and an additional 9 genes at FDR < 0.2, with many of these previously identified as involved in specific neurodevelopmental disorders. Point estimates of the liability conferred by de novo variants on the X are similar in females and males, with both sexes' estimates elevated >20% above the corresponding autosomal values. We also develop a general theory of how X-linked variation of any additive or non-additive effect influences liability and describe its implications for prevalence. Using this theory and our empirical results, we show how genetic variation on the X could contribute to the sex-differential prevalence of ASD.

Motivation Gene-damaging mutations are highly informative for studies seeking to discover genes underlying developmental disorders. Traditionally, these de novo variants are recognized by evaluating high-quality DNA sequence from affected offspring and parents. However, when parental sequence is unavailable, methods are required to infer de novo status and use this inference for association studies.Results We use data from autism spectrum disorder to illustrate and evaluate methods. Separating de novo from rare inherited variants is challenging because the latter are far more common. Using a classifier for unbalanced data and variants of known inheritance class, we build an inheritance model and then a de novo score for variants when parental data are missing. Next, we propose a new Random Draw (RD) model to use this score for gene discovery. Built into an existing inferential framework, RD produces a more powerful gene-based association test and controls the false discovery rate.Availability and implementation Codes are available at Github (https://github.com/HaeunM/TADA-RD) and Zenodo (DOI: https://doi.org/10.5281/zenodo.18531769).

Autism spectrum disorder is a heritable neurodevelopmental condition affecting approximately 3% of children that presents with core behavioral features and a range of possible comorbidities, including intellectual disability. While common variants contribute substantially to autism liability, the discovery of specific autism-associated genes has largely been driven by studies of rare and de novo variants. Many of these genes are also linked with broadly defined developmental disorders, but their involvement in other conditions has not been mapped at scale. Here, we analyze autosomal rare coding variation from 62,429 individuals with autism from research and clinical cohorts to identify 253 autism-associated genes at an estimated false discovery rate < 0.001. We cluster them based on association evidence from large-scale studies of developmental disorders, schizophrenia, bipolar disorder, and epilepsy, generating six clusters of genes with differing biological pathway enrichments and patterns of comorbidities. Investigating rare variant associations in the population using the UK Biobank and All of Us, we identify autism-associated genes displaying pleiotropy across physiological systems. In addition, we report 497 genes impacting development in a meta-analysis with 26,109 published developmental disorders samples. Collectively drawing upon data from over 1.5 million individuals, our study finds that rare variants across hundreds of genes contribute to autism with variable phenotypic outcomes.

Motivation: Kinases regulate a multitude of protein functions, and their dysregulation is pivotal for many human diseases. Direct measurement of kinase activity, however, is often challenging; therefore, inferring activity from the behavior of their substrates is a widely adopted strategy. Nonetheless, traditional methods typically oversimplify the underlying network, ignoring that any particular substrate can be phosphorylated by multiple kinases. Results: We present LIKA, a likelihood-based framework for inferring kinase activity from phosphoproteomic data. By modeling the many-to-many structure of kinase-substrate interactions, LIKA achieves high efficiency, even with limited data, while capturing network complexity. Simulation and cell line analyses confirm the robustness and accuracy of LIKA. Importantly, analysis of a phosphoproteomic dataset from schizophrenia and control subjects reveals novel dysregulated kinases. Availability and Implementation: The implementation code and publicly available data are provided at: https://github.com/lujingz/LIKA.

The past decade has seen remarkable progress in identifying genes that, when impacted by deleterious coding variation, confer high likelihood for autism spectrum disorder (ASD), intellectual disability and other associated developmental disorders. However, most underlying gene discovery efforts have focused on individuals of European ancestry, limiting insights into genetic liability across diverse populations. To help address this, the Genomics of Autism in Latin American Ancestries (GALA) Consortium was formed, presenting here the largest sequencing study of autism in Latin American individuals (n > 15,000, including 4,717 participants with an ASD diagnosis). We identified 35 genome-wide significant (false discovery rate < 0.05) autism-associated genes, with substantial overlap with findings from European cohorts, and highly constrained genes showing consistent signal across populations. The results provide support for emerging (for example, MARK2, YWHAG, PACS1, RERE, SPEN, GSE1, GLS, TNPO3 and ANKRD17) and established autism genes and for the utility of genetic testing approaches for deleterious variants in individuals from diverse backgrounds; the results also demonstrate the ongoing need for more inclusive genetic research and testing. We conclude that the biology of autism is consistent across populations, with no detectable influence of ancestry.

BACKGROUND:Autism spectrum disorder (ASD) has a complex inheritance pattern and is more common in males. Etiologic models suggest that most ASD risk is transmitted through common and rare de novo genetic variation. It has been hypothesized that rare variation could be inherited and therefore contribute to the overall risk burden in subsequent generations, especially through female lineage in disorders with male-skewed sex ratios. Here, we tested this hypothesis using multigeneration information on paternal age, because burden of de novo mutations has been linked to paternal age, and there is a well-established association between older age of fathers and ASD. METHODS:We analyzed combined data from Sweden's, Denmark's, and Finland's national registers, totaling 12.6 million family members, including information about parental ages at the time of birth of offspring in 2 generations and ASD diagnosis in the third generation. RESULTS:Among the 1,808,892 children in the third generation, 23,397 (1.29%) were diagnosed with ASD. Increased paternal age at the time of birth of a daughter was associated with increased risk of ASD in the daughter's own offspring. Increased paternal age at the time of birth of a son was not associated with increased ASD risk in the son's offspring, nor was older maternal age in the first or second generations. We observed that young maternal age at birth of a son or a daughter was associated with ASD risk in their offspring. CONCLUSIONS:Collectively, our results suggest that etiologic risk factors for ASD could extend over multiple generations through different underlying mechanisms, suggesting new directions for research on genetic and nongenetic risk factors.

PACS1 syndrome is a neurodevelopmental disorder (NDD) resulting from a unique de novo p.R203W variant in Phosphofurin Acidic Cluster Sorting protein 1 (PACS1). PACS1 encodes a multifunctional sorting protein required for localizing furin to the trans-Golgi network. Although few studies have started to investigate the impact of the PACS1 p.R203W variant, the mechanisms by which the variant affects neurodevelopment are still poorly understood. In recent years, autism spectrum disorder (ASD) patient-derived brain organoids have been increasingly used to identify pathogenic mechanisms and possible therapeutic targets. While most of these studies evaluate the mechanisms by which ASD-risk genes affect the transcriptome, studies considering the proteome are limited. Here, we examine the effect of PACS1 p.R203W on the proteomic landscape of brain organoids using tandem mass tag (TMT) mass-spectrometry. Time series analysis between PACS1(+/+) and PACS1(+/R203W) organoids uncovered several proteins with dysregulated abundance or phosphorylation status, including known PACS1 interactors. Although we observed low overlap between proteins with altered expression and phosphorylation, the resulting dysregulated processes converged. The presence of the PACS1 p.R203W variant accelerated the emergence of proteins related to synaptogenesis and impaired vesicle loading and recycling. The earlier presence of these proteins and their related processes could lead to defective and/or incomplete synaptic function. Key dysregulated proteins observed in PACS1(+/R203W) organoids have been associated with several neurological diseases, and many are classified as NDD-causative and ASD-risk genes. Our results highlight that proteomic analyses not only enhance our understanding of general NDD mechanisms by complementing transcriptomic studies, but could also uncover additional targets, and therefore facilitate therapy development.

New insights into genetic etiological factors underlying autism and related neurodevelopmental and neuropsychiatric conditions have progressed rapidly, driven by accelerating data aggregation and analytic innovations. Large-scale sequencing of rare coding variation has enabled robust gene discovery and deeper insights into biological mechanisms underlying human development and cognition. Here, we report the largest-to-date analysis of rare variants in autism, encompassing 62,470 individuals diagnosed with autism, including 38,545 probands with parental data from complete families. By integrating de novo and inherited data across single-nucloetide and copy number variation in this cohort via the TADA Bayesian model, we identify 257 genes robustly associated with autism at a false discovery rate (FDR) < 0.001. Not only do these genes recapitulate strong enrichment in pathways such as those involved in chromatin remodeling, development, and synaptic communication/signaling, but many of them have also recently been shown to be significantly associated in studies across a range of neuropsychiatric disorders.We sought to contextualize these findings in the broader landscape of neuropsychiatric genetics by systematically aggregating our results gene and pathway level findings from large-scale studies of schizophrenia (SCHEMA; 24,248 cases, 97,322 controls), epilepsy (Epi25K; 20,979 cases, 33,444 controls), and bipolar disorder (BIPEX; 13,933 cases, 14,422 controls). Burden heritability regression reveals that autism harbors the greatest rare variant heritability (>3%), followed by schizophrenia and epilepsy (1–2%). We observe moderate genetic correlation between autism and each of schizophrenia, epilepsy, and bipolar disorder based on rare variants (∼0.2), while the highest rare variant genetic correlation is between schizophrenia and epilepsy (∼0.5), and in schizophrenia with bipolar disorder (∼0.4). Partitioning rare variant heritability reveals that 25% of rare variant heritability in autism resides in the 257 autism-associated genes, while the same genes account for ∼20% of rare variant heritability in epilepsy, but less than 10% in schizophrenia and bipolar disorder, suggesting both shared and distinct pathways of disruption. At a gene-set level, the autism associated genes are significantly more likely to also be associated with schizophrenia (Odds ratio [OR]=10.8, p=1.5e-14) and epilepsy (OR=11.6, p=1.1e-11), with weaker enrichment in bipolar disorder (OR=2.8, p=0.094). Finally, autism genes in chromatin, development, and synaptic signaling pathways are significantly enriched for genes associated with schizophrenia (OR=1.76, p=7.6e-2) and epilepsy (OR=2.22, p=3.2e-4).Our findings highlight a core set of highly penetrant genes with impact across autism and neuropsychiatric phenotypes, while also revealing disorder-specific genetic architecture differences. Additional efforts to integrate these gene-level discoveries with developmental expression patterns, cell-type specificity, and functional networks will further inform functional mechanisms for rare variant risk across diagnostic boundaries.

The molecular basis of phenotypes is often explored by contrasting gene expression from relevant tissue taken from individuals classified into phenotypic extremes, such as affected versus unaffected individuals. Analysis of this differential expression (DE) typically identifies many genes of interest. However, it is not clear which genes differ between extremes because they alter phenotypic liability and which show differences as a result of the extreme phenotype itself. We propose a formal model to distinguish between genes that are upstream and “cause” differential expression versus those for which differential expression is a result of the initial manifestation of phenotype. Relying on two sets of p -values, one from differential expression analysis and one from gene-specific association with phenotype (AP), and a gene coexpression or other gene-based network that serves as a bridge, our method identifies communities of genes more likely upstream or downstream of the phenotype. Our method consists of three major steps: 1) gene network construction, 2) evaluation of DE and AP signal within the network to infer hidden states, and 3) detection of gene communities. We apply our method to data that were generated to assess the biological basis of autism spectrum disorder (ASD) and Alzheimer’s disease (AD). Our results highlight neuronal and synaptic biology as being upstream of ASD, whereas downstream processes are all non-neuronal. For AD, our results are consistent with existing hypotheses; yet, they also lend support for a recent unifying hypothesis involving cofilin/actin biology.

Importance:Obsessive-compulsive disorder (OCD) affects 2-3% of the population with often disabling obsessions and compulsions. Despite its high heritability, genetic studies of OCD have lagged other psychiatric disorders, particularly in understanding the role of rare genetic variants. Objective:To identify rare coding genetic variants contributing to OCD risk and examine genetic overlap with chronic tic disorders (CTD) and other psychiatric conditions. Design:Family-based and case-control whole-exome sequencing (WES) study. Settings:WES data were aggregated from 11 independent cohorts across Sweden, the United States, and the United Kingdom. Participants:A total of 47,194 individuals were available, and 44,089 passed quality control for analysis. The final sample included 6,071 individuals with OCD, comprising 1,202 probands from family-based trios and 4,869 cases, and 38,018 controls. Exposures:Rare damaging coding variants identified by WES. Main Outcomes and Measures:Identification of OCD risk genes through rare variant analyses, meta-analysis with CTD data, gene-set enrichment analyses, and evaluation of cross-disorder genetic overlap using curated gene sets. Results:The analysis provided an estimate of approximately 470 autosomal genes contributing to OCD risk through rare genetic variation. CHD8 reached genome-wide significance (q < 0.05). Meta-analysis with CTD data revealed additional risk genes, including CELSR3 (q < 0.05), QRICH1, and WWC1 (q < 0.1). We observed significant genetic overlap between OCD, autism spectrum disorder (ASD), and developmental delay: 33% of ASD genes with FDR < 0.1 showed association with OCD (p < 0.001), and 36% showed possible associations in the shared OCD-CTD genetic architecture (p < 0.001), but minimal rare-variant overlap with bipolar disorder and schizophrenia risk genes. We also found that CHD8-regulated genes were enriched for both rare and common variant associations with OCD. Conclusions and Relevance:In this largest study to date of rare coding variation in OCD, we confirm CHD8 as the first genome-wide significant rare-variant risk gene, show that genes that are targets of CHD8 can carry rare and common variant risk for OCD, and identify multiple additional genes and pathways contributing to risk. Taken together, the findings show that OCD shares substantially greater genetic overlap with neurodevelopmental conditions than with adult-onset psychiatric disorders, refining the developmental framework of OCD and informing future mechanistic and clinical research.

Elevated maternal pre-pregnancy body mass index (BMI) has been suggested to increase risk of offspring autism spectrum disorder (ASD) but evidence is mixed across heterogeneous studies and robust estimates spanning the full BMI range are lacking. This study examined the association between maternal BMI and offspring ASD in a harmonized, two-nation study and across the full BMI range. We included all singleton children born in Denmark 2004–2018 and Sweden 1998–2019 to parents of Nordic origin (n = 2,072,445), with follow-up from age 2 until 31 December 2021, or 2022, respectively. Maternal BMI recorded at the first antenatal visit was obtained from the Swedish and Danish Medical Birth Registers and was analyzed as a continuous variable and in World Health Organization-defined categories of underweight (BMI < 18.5), normal weight (18.5–24.9), overweight (25–29.9), obese class I (30–34.9), and obese class II–III (≥ 35). The relative risk of ASD was estimated as hazard ratios (HR) from Cox regression models, adjusted for birth year and parental age, educational level, income, and psychiatric history at time of childbirth, using data from national health and population registers. Both country-specific and pooled analyses were conducted. Subgroup and sensitivity analyses, including a sibling comparison, were performed to address the specificity and robustness of findings. A total of 58,416 (2.8

Background Autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) are heterogeneous neurodevelopmental disorders with high heritability and frequent co-occurrence. Our previous work on the initial iPSYCH exomes (Satterstrom et al., 2019) suggested a similar burden of rare protein-truncating variants (PTVs) across ASD and ADHD and identified MAP1A as a shared risk gene implicated by rare PTVs in both disorders. To build upon these findings, we aimed to 1) expand our gene discovery analysis using an updated dataset with nearly twice the sample size from the latest iPSYCH exomes, 2) quantify the burden heritability attributable to rare coding variants in ASD and ADHD, and 3) evaluate the burden genetic correlation between two disorders. Methods We analyzed exomes of 25,208 individuals from iPSYCH, comprising 7,119 diagnosed with ASD alone (ASD-only), 5,598 with ADHD alone (ADHD-only), 3,794 diagnosed with both conditions (ASD+ADHD), and 8,697 controls. Multivariate Poisson regression models were applied to systematically assess rare variant burdens in various gene sets across the three case groups, further stratifying by the presence or absence of intellectual disability (ID). We employed c-alpha tests to compare the distribution of rare deleterious variants between ASD-only and ADHD-only. We performed burden heritability regression analyses to estimate the burden heritability of ASD and ADHD, respectively, and to measure their burden genetic correlation. For gene discovery, we combined individuals diagnosed with ASD and/or ADHD into a single case group, included non-psychiatric non-Finnish European exome subset of gnomAD as external controls, and applied Fisher’s exact test to identify genes reaching exome-wide significance. Results Consistent with our previous findings, all three case groups demonstrated comparable elevated burdens of class I variants - including rare PTVs and highly deleterious missense variants (AlphaMissense ≥ 0.98 and MPC ≥ 2) in constrained genes compared to controls (ASD-only: OR = 1.49, 95% CI [1.40, 1.58]; ADHD-only: OR = 1.40, 95% CI [1.31, 1.50]; ASD+ADHD: OR = 1.46, 95% CI [1.35, 1.57]). C-alpha tests indicated no significant differences in the distribution of these variants between ASD-only and ADHD-only (P= 0.40), whereas significant differences were observed when comparing each group to controls. Burden heritability estimates of class I variants were 1.8% (s.e. = 0.4%) for ASD and 3.2% (s.e. = 0.7%) for ADHD on the liability scale. The burden genetic correlation between the two disorders was 0.46 (s.e. = 0.17), aligning closely with previously reported common-variant genetic correlation (0.42, s.e. = 0.05; Demontis et al., 2023). In gene discovery, we identified eight exome-wide significant genes associated with both disorders, including MAP1A (the first cross-disorder gene previously identified) and seven new risk genes: five previously implicated in ASD, developmental delay, and neurodevelopmental disorders; one strong candidate gene for ASD; and one novel gene not previously linked to either disorder. Discussion Our findings underscore a substantial shared genetic architecture involving rare coding variants between ASD and ADHD, reinforcing and expanding on earlier research. Moving forward, we aim to explore the distinct genetic risks specific to each disorder and to conduct sex-stratified analyses to uncover potential sex-specific genetic differences. The results will be presented at the conference.

The past decade has seen remarkable progress in identifying genes that, when impacted by deleterious coding variation, confer high risk for autism spectrum disorder (ASD), intellectual disability, and other developmental disorders. However, most underlying gene discovery efforts have focused on individuals of European ancestry, limiting insights into genetic risks across diverse populations. To help address this, the Genomics of Autism in Latin American Ancestries Consortium (GALA) was formed, presenting here the largest sequencing study of ASD in Latin American individuals (n>15,000). We identified 35 genome-wide significant (FDR < 0.05) ASD risk genes, with substantial overlap with findings from European cohorts, and highly constrained genes showing consistent signal across populations. The results provide support for emerging (e.g., MARK2, YWHAG, PACS1, RERE, SPEN, GSE1, GLS, TNPO3, ANKRD17) and established ASD genes, and for the utility of genetic testing approaches for deleterious variants in diverse populations, while also demonstrating the ongoing need for more inclusive genetic research and testing. We conclude that the biology of ASD is universal and not impacted to any detectable degree by ancestry.

Autism spectrum disorder (ASD) and attention deficit hyperactivity disorder (ADHD) are heterogeneous neurodevelopmental disorders with high heritability and frequent co-occurrence. Our previous work on the first phase of iPSYCH exomes (Satterstrom et al., 2019) suggested a similar burden of rare protein-truncating variants (PTVs) across ASD and ADHD and identified MAP1A as a shared risk gene implicated by rare PTVs in both disorders. This study aims to 1) extend these findings, employing a significantly larger iPSYCH exome dataset for gene discovery, 2) estimate the burden heritability explained by rare coding variants in ASD and ADHD, and 3) assess the burden genetic correlation between the two disorders.We analyzed exomes of 25,208 individuals from iPSYCH, encompassing 7,119 individuals diagnosed with ASD alone (ASD-only), 5,598 with ADHD alone (ADHD-only), 3,794 with both ASD and ADHD (ASD+ADHD), and 8,697 controls. We used multivariate Poisson regression models to systematically evaluate rare variant burdens in different gene sets across the three case groups and controls, stratified further by the presence or absence of intellectual disability (ID). The gene sets included all genes, genes intolerant to loss-of-function variants (pLI > 0.9), and gene sets associated with different disorders including ID, ASD, ADHD, schizophrenia, and a broader group of neurodevelopmental disorders. We applied c-alpha tests to assess whether the distribution of rare deleterious variants differs between ASD and ADHD. We employed burden heritability regression to estimate the burden heritability of ASD and ADHD, respectively, and the burden genetic correlation between the two disorders. For gene discovery, we combined individuals diagnosed with ASD and/or ADHD into a single case group and applied TADA+ to integrate with family data and Swedish PAGES case-control data from a recent large-scale ASD rare variant study (Fu et al., 2022).We observed similar burdens of class I variants including rare PTVs and rare deleterious missense variants (MPC > 3) in constrained genes across the three case groups, while they all showed a significant excess compared to controls: OR = 1.35, 95% CI = [1.26, 1.45] for ASD-only; OR = 1.35, CI = [1.25, 1.45] for ADHD-only; and OR = 1.39, CI = [1.28, 1.52] for ASD+ADHD. The c-alpha tests indicated no significant differences in the distribution of class I variants in constrained genes between ASD-only and ADHD-only groups (P= 0.39) while, when comparing the case groups to controls, significant differences were observed. The burden heritability of class I variants on the liability scale was estimated to 1.87% (SE = 0.51%) for ASD and 2.42% (s.e. = 0.72%) for ADHD. The class I variant burden genetic correlation between ASD and ADHD was 0.31 (s.e. = 0.26), which approximates the point estimate of their common-variant genetic correlation of 0.42 (s.e. = 0.05) (Demontis et al., 2023).Our findings suggest substantial sharing of rare variant risk between ASD and ADHD, reinforcing the results of our earlier work (Satterstrom et al., 2019). This motivated us to merge individuals diagnosed with ASD and/or ADHD into a single group to enhance the discovery of rare variant risk genes shared between the disorders. This gene discovery analysis is ongoing, and the results will be presented at the conference.

INTRODUCTION Individuals with Alzheimer's disease (AD) commonly experience neuropsychiatric symptoms of psychosis (AD+P) and/or affective disturbance (depression, anxiety, and/or irritability, AD+A). This study's goal was to identify the genetic architecture of AD+P and AD+A, as well as their genetically correlated phenotypes. METHOD SGenome-wide association meta-analysis of 9988 AD participants from six source studies with participants characterized for AD+P AD+A, and a joint phenotype (AD+A+P). RESULTS AD+P and AD+A were genetically correlated. However, AD+P and AD+A diverged in their genetic correlations with psychiatric phenotypes in individuals without AD. AD+P was negatively genetically correlated with bipolar disorder and positively with depressive symptoms. AD+A was positively correlated with anxiety disorder and more strongly correlated than AD+P with depressive symptoms. AD+P and AD+A+P had significant estimated heritability, whereas AD+A did not. Examination of the loci most strongly associated with the three phenotypes revealed overlapping and unique associations. DISCUSSION AD+P, AD+A, and AD+A+P have both shared and divergent genetic associations pointing to the importance of incorporating genetic insights into future treatment development.

Polygenic scores (PGSs) are quantitative metrics for predicting phenotypic values, such as human height or disease status. Some PGS methods require only summary statistics of a relevant genome-wide association study (GWAS) for their score. One such method is Lassosum, which inherits the model selection advantages of Lasso to select a meaningful subset of the GWAS single-nucleotide polymorphisms as predictors from their association statistics. However, even efficient scores like Lassosum, when derived from European-based GWASs, are poor predictors of phenotype for subjects of non-European ancestry; that is, they have limited portability to other ancestries. To increase the portability of Lassosum, when GWAS information and estimates of linkage disequilibrium are available for both ancestries, we propose Joint-Lassosum (JLS). In the simulation settings we explore, JLS provides more accurate PGSs compared to other methods, especially when measured in terms of fairness. In analyses of UK Biobank data, JLS was computationally more efficient but slightly less accurate than a Bayesian comparator, SDPRX. Like all PGS methods, JLS requires selection of predictors, which are determined by data-driven tuning parameters. We describe a new approach to selecting tuning parameters and note its relevance for model selection for any PGS. We also draw connections to the literature on algorithmic fairness and discuss how JLS can help mitigate fairness-related harms that might result from the use of PGSs in clinical settings. While no PGS method is likely to be universally portable, due to the diversity of human populations and unequal information content of GWASs for different ancestries, JLS is an effective approach for enhancing portability and reducing predictive bias.

The fields of autism and neurodevelopmental disorder (NDD) genetics are rapidly advancing. Catalyzed by the power of large cohorts and integration of all classes of de novo and inherited protein-coding variation, dozens of genes have emerged to harbor variants that confer high relative risk for autism, and hundreds of genes have been associated with NDDs more broadly. Through examination of protein-truncating variants (PTVs), predicted damaging missense variation, and copy number variants (CNVs), our prior analyses have begun to map the allelic diversity of perturbations within 72 autism-associated genes and 373 genes associated with NDDs, finding intriguing evidence of genes with significantly higher mutation rates and differences in the distribution of clinical phenotypes in autism compared to NDD (Fu et al., 2022; Satterstrom et al., 2020). Despite this progress, cohort sizes remain insufficient for disentangling the shared and distinct genetic architectures of autism, NDDs, and other neuropsychiatric conditions, as well as associating genes with more subtle impacts on neurodevelopment.To advance these boundaries, we present the largest to-date study of rare coding variants, consisting of 62,013 autistic individuals, including 38,088 probands and 9,567 unaffected siblings from complete trio and quartet families, respectively, and 23,925 additional autism cases without parental information contrasted against 26,931 controls. By aggregating across the Autism Sequencing Consortium (ASC), the Simons Simplex Collection (SSC), the Simons Foundation Powering Autism Research (SPARK), and individuals from a leading diagnostic laboratory (GeneDx), this dataset totals almost 200,000 individuals, nearly a three-fold increase over prior studies. When we stratified the clinically-referred GeneDx autistic probands by co-occurring DD/ID status, we found synonymous, missense, and PTV de novo mutation rates in autism probands without DD/ID from GeneDx that were nearly identical to individuals ascertained for a diagnosis of autism in the ASC, SSC, and SPARK research studies (0.296 vs 0.294, 0.767 vs 0.763, and 0.141 vs 0.145 respectively), while GeneDx autism probands with DD/ID exhibited mutation rates similar to those observed in previous research studies of DD.Further analyses of these data solidified previous observations of significant enrichment of de novo PTVs among autism probands of 3x compared to siblings among the genes most intolerant to PTVs in the human genome (i.e., lowest decile of LOEUF from gnomAD). We have also incorporated Alpha Missense (AM) pathogenicity estimates to complement our prior MPC scores for predicting damaging missense variation and identifying de novo missense variants acting with effect sizes comparable to de novo PTVs in constrained genes, with analysis of regional missense constraint within genes ongoing. We further leveraged the TADA Bayesian statistical method to jointly model these data in a single unified framework, leveraging genetic information across rare PTVs, damaging missense variants, and CNVs. This approach discovered hundreds of genes associated with autism, where we observe a steadily increasing contribution of variant classes other than de novo PTVs in newly associated genes. Analyses are ongoing to understand the gene networks, developmental timing, and biological functions by which these genes exert their influence on phenotypic manifestations of autism and related neuropsychiatric disorders.