Heritability enrichment of specifically expressed genes identifies disease-relevant tissues and cell types

Omics Approaches to Investigate the Pathogenesis of Suicide

Suicide is the second leading cause of death in U.S. adolescents and young adults and is generally associated with a psychiatric disorder. Suicidal behavior has a complex etiology and pathogenesis. Moderate heritability suggests genetic causes. Associations between childhood and recent life adversity indicate contributions from epigenetic factors. Genomic contributions to suicide pathogenesis remain largely unknown. This article is based on a workshop held to design strategies to identify molecular drivers of suicide neurobiology that would be putative new treatment targets. The panel determined that while bulk tissue studies provide comprehensive information, single-nucleus approaches that identify cell type-specific changes are needed. While single-nuclei techniques lack information on cytoplasm, processes, spines, and synapses, spatial multiomic technologies on intact tissue detect cell alterations specific to brain tissue layers and subregions. Because suicide has genetic and environmental drivers, multiomic approaches that combine cell type-specific epigenome, transcriptome, and proteome provide a more complete picture of pathogenesis. To determine the direction of effect of suicide risk gene variants on RNA and protein expression and how these interact with epigenetic marks, single-nuclei and spatial multiomics quantitative trait loci maps should be integrated with whole-genome sequencing and genome-wide association databases. The workshop concluded with a recommendation for the formation of an international suicide biology consortium that will bring together brain banks and investigators with expertise in cutting-edge omics technologies to delineate the biology of suicide and identify novel potential treatment targets to be tested in cellular and animal models for drug and biomarker discovery to guide suicide prevention.

Exploring the shared genetic basis of major depressive disorder and frailty

BACKGROUND

Major depressive disorder (MDD) and frailty impose substantial health and economic burdens. MDD is recognized as a significant risk factor for frailty, but the genetic associations between these conditions remain unclear. This study investigates the genetic correlation, shared pleiotropic loci, causal relationships, and comorbid genes between MDD and frailty.

METHODS

The genetic correlation between MDD and frailty was assessed using linkage disequilibrium score regression (LDSC) based on data from genome-wide association studies (GWAS). A detailed analysis was performed to identify shared pleiotropic loci and causal relationships through cross-phenotype association tests and Mendelian randomization. Additionally, tissue enrichment analysis was conducted using stratified LDSC, gene-based associations with both conditions were assessed using Multimarker Analysis of Genomic Annotation (MAGMA), and pathway analysis of comorbid genes was performed using the g: GOSt tool.

RESULTS

Our findings revealed a significant positive genetic correlation between MDD and frailty (rg = 0.65, P = 1.49E-219). We identified 57 shared risk SNPs between the two conditions, including 6 novel SNPs. Mendelian randomization analyses indicated robust causal effects of MDD on frailty and vice versa. Furthermore, we observed tissue-specific heritability enrichment in 9 brain tissues. By combining MAGMA and CPASSOC analyses, we identified 10 comorbid genes associated with both MDD and frailty, primarily involved in synapse formation, modulation, plasticity, and desaturase activity.

CONCLUSION

This study provides strong evidence for a shared genetic basis between MDD and frailty. The identification of comorbid genes offers new insights into the mechanisms underlying the relationship between these conditions.

Cross-ancestry analysis of brain QTLs enhances interpretation of schizophrenia genome-wide association studies

Research on brain expression quantitative trait loci (eQTLs) has illuminated the genetic underpinnings of schizophrenia (SCZ). Yet most of these studies have been centered on European populations, leading to a constrained understanding of population diversities and disease risks. To address this gap, we examined genotype and RNA-seq data from African Americans (AA, n = 158), Europeans (EUR, n = 408), and East Asians (EAS, n = 217). When comparing eQTLs between EUR and non-EUR populations, we observed concordant patterns of genetic regulatory effect, particularly in terms of the effect sizes of the eQTLs. However, 343,737 cis-eQTLs linked to 1,276 genes and 198,769 SNPs were found to be specific to non-EUR populations. Over 90% of observed population differences in eQTLs could be traced back to differences in allele frequency. Furthermore, 35% of these eQTLs were notably rare in the EUR population. Integrating brain eQTLs with SCZ signals from diverse populations, we observed a higher disease heritability enrichment of brain eQTLs in matched populations compared to mismatched ones. Prioritization analysis identified five risk genes (SFXN2, VPS37B, DENR, FTCDNL1, and NT5DC2) and three potential regulatory variants in known risk genes (CNNM2, MTRFR, and MPHOSPH9) that were missed in the EUR dataset. Our findings underscore that increasing genetic ancestral diversity is more efficient for power improvement than merely increasing the sample size within single-ancestry eQTLs datasets. Such a strategy will not only improve our understanding of the biological underpinnings of population structures but also pave the way for the identification of risk genes in SCZ.

Uncovering cell-type-specific immunomodulatory variants and molecular phenotypes in COVID-19 using structurally resolved protein networks

Immunomodulatory variants that lead to the loss or gain of specific protein interactions often manifest only as organismal phenotypes in infectious disease. Here, we propose a network-based approach to integrate genetic variation with a structurally resolved human protein interactome network to prioritize immunomodulatory variants in COVID-19. We find that, in addition to variants that pass genome-wide significance thresholds, variants at the interface of specific protein-protein interactions, even though they do not meet genome-wide thresholds, are equally immunomodulatory. The integration of these variants with single-cell epigenomic and transcriptomic data prioritizes myeloid and T cell subsets as the most affected by these variants across both the peripheral blood and the lung compartments. Of particular interest is a common coding variant that disrupts the OAS1-PRMT6 interaction and affects downstream interferon signaling. Critically, our framework is generalizable across infectious disease contexts and can be used to implicate immunomodulatory variants that do not meet genome-wide significance thresholds.

Multivariate genomic analysis of 5 million people elucidates the genetic architecture of shared components of the metabolic syndrome

Metabolic syndrome (MetS) is a complex hereditary condition comprising various metabolic traits as risk factors. Although the genetics of individual MetS components have been investigated actively through large-scale genome-wide association studies, the conjoint genetic architecture has not been fully elucidated. Here, we performed the largest multivariate genome-wide association study of MetS in Europe (nobserved = 4,947,860) by leveraging genetic correlation between MetS components. We identified 1,307 genetic loci associated with MetS that were enriched primarily in brain tissues. Using transcriptomic data, we identified 11 genes associated strongly with MetS. Our phenome-wide association and Mendelian randomization analyses highlighted associations of MetS with diverse diseases beyond cardiometabolic diseases. Polygenic risk score analysis demonstrated better discrimination of MetS and predictive power in European and East Asian populations. Altogether, our findings will guide future studies aimed at elucidating the genetic architecture of MetS.

Dissecting the polygenic contribution of attention-deficit/hyperactivity disorder and autism spectrum disorder on school performance by their relationship with educational attainment

Attention-deficit/hyperactivity disorder (ADHD) and autism spectrum disorders (ASD) are strongly associated with educational attainment (EA), but little is known about their genetic relationship with school performance and whether these links are explained, in part, by the genetic liability of EA. Here, we aim to dissect the polygenic contribution of ADHD and ASD to school performance, early manifestation of psychopathology and other psychiatric disorders and related traits by their relationship with EA. To do so, we tested the association of polygenic scores for EA, ADHD and ASD with school performance, assessed whether the contribution of the genetic liability of ADHD and ASD to school performance is influenced by the genetic liability of EA, and evaluated the role of EA in the genetic overlap between ADHD and ASD with early manifestation of psychopathology and other psychiatric disorders and related traits in a sample of 4,278 school-age children. The genetic liability for ADHD and ASD dissected by their relationship with EA show differences in their association with school performance and early manifestation of psychopathology, partly mediated by ADHD and ASD symptoms. Genetic variation with concordant effects in ASD and EA contributes to better school performance, while the genetic variation with discordant effects in ADHD or ASD and EA is associated with poor school performance and higher rates of emotional and behavioral problems. Our results strongly support the usage of the genetic load for EA to dissect the genetic and phenotypic heterogeneity of ADHD and ASD, which could help to fill the gap of knowledge of mechanisms underlying educational outcomes.

Genetic analysis of key agronomic traits of local sheep breeds in Xinjiang, China

The formation of sheep (Ovis aries) breeds is influenced by different ecological environments and populations with different living habits, resulting in the development of germplasm resources with stable genetic key agronomic traits. Thus, investigating the genetic mechanisms behind various agronomic traits can enhance the conservation and utilization of diverse sheep breeds. Here, we explored the sheep variome and selection signatures using the Ovine Infinium HD SNP BeadChip (600 K SNPs) from 23 sheep breeds, comprising a total of 1215 sheep. The genetic mechanisms of wool quality and tail morphology were analyzed by selective sweep and genome-wide association study. Based on the results of within-population selective sweep analysis, we performed gene network analysis and divided them into 6 gene communities. We identified genetic regions containing genes linked to sheep wool and tail, which have been and may continue to be important targets for breeding and selection. Furthermore, our results revealed the expression profiles of genes in these regions across different biological systems. Our study provides insights into categorizing sheep breeds into distinct gene communities, as well as references for constructing genetic network pathways related to key agronomic traits in sheep and other domestic animals.

Temporally distinct 3D multi-omic dynamics in the developing human brain

The human hippocampus and prefrontal cortex play critical roles in learning and cognition1,2, yet the dynamic molecular characteristics of their development remain enigmatic. Here we investigated the epigenomic and three-dimensional chromatin conformational reorganization during the development of the hippocampus and prefrontal cortex, using more than 53,000 joint single-nucleus profiles of chromatin conformation and DNA methylation generated by single-nucleus methyl-3C sequencing (snm3C-seq3)3. The remodelling of DNA methylation is temporally separated from chromatin conformation dynamics. Using single-cell profiling and multimodal single-molecule imaging approaches, we have found that short-range chromatin interactions are enriched in neurons, whereas long-range interactions are enriched in glial cells and non-brain tissues. We reconstructed the regulatory programs of cell-type development and differentiation, finding putatively causal common variants for schizophrenia strongly overlapping with chromatin loop-connected, cell-type-specific regulatory regions. Our data provide multimodal resources for studying gene regulatory dynamics in brain development and demonstrate that single-cell three-dimensional multi-omics is a powerful approach for dissecting neuropsychiatric risk loci.

Discovery and prioritization of genetic determinants of kidney function in 297,355 individuals from Taiwan and Japan

Current genome-wide association studies (GWAS) for kidney function lack ancestral diversity, limiting the applicability to broader populations. The East-Asian population is especially under-represented, despite having the highest global burden of end-stage kidney disease. We conducted a meta-analysis of multiple GWASs (n = 244,952) on estimated glomerular filtration rate and a replication dataset (n = 27,058) from Taiwan and Japan. This study identified 111 lead SNPs in 97 genomic risk loci. Functional enrichment analyses revealed that variants associated with F12 gene and a missense mutation in ABCG2 may contribute to chronic kidney disease (CKD) through influencing inflammation, coagulation, and urate metabolism pathways. In independent cohorts from Taiwan (n = 25,345) and the United Kingdom (n = 260,245), polygenic risk scores (PRSs) for CKD significantly stratified the risk of CKD (p < 0.0001). Further research is required to evaluate the clinical effectiveness of PRSCKD in the early prevention of kidney disease.

Comparative single-cell multiome identifies evolutionary changes in neural progenitor cells during primate brain development

Understanding the cellular and genetic mechanisms driving human-specific features of cortical development remains a challenge. We generated a cell-type resolved atlas of transcriptome and chromatin accessibility in the developing macaque and mouse prefrontal cortex (PFC). Comparing with published human data, our findings demonstrate that although the cortex cellular composition is overall conserved across species, progenitor cells show significant evolutionary divergence in cellular properties. Specifically, human neural progenitors exhibit extensive transcriptional rewiring in growth factor and extracellular matrix (ECM) pathways. Expression of the human-specific progenitor marker ITGA2 in the fetal mouse cortex increases the progenitor proliferation and the proportion of upper-layer neurons. These transcriptional divergences are primarily driven by altered activity in the distal regulatory elements. The chromatin regions with human-gained accessibility are enriched with human-specific sequence changes and polymorphisms linked to intelligence and neuropsychiatric disorders. Our results identify evolutionary changes in neural progenitors and putative gene regulatory mechanisms shaping primate brain evolution.

GWAS shows the genetics behind cell-free DNA and highlights the importance of p.Arg206Cys in DNASE1L3 for non-invasive testing

The properties of cell-free DNA (cfDNA) are intensely studied for their potential as non-invasive biomarkers. We explored the effect of common genetic variants on the concentration and fragmentation properties of cfDNA using a genome-wide association study (GWAS) based on low-coverage whole-genome sequencing data of 140,000 Dutch non-invasive prenatal tests (NIPTs). Our GWAS detects many genome-wide significant loci, functional enrichments for phagocytes, liver, adipose tissue, and macrophages, and genetic correlations with autoimmune and cardiovascular disease. A common (7%) missense variant in DNASE1L3 (p.Arg206Cys) strongly affects all cfDNA properties. It increases the size of fragments, lowers cfDNA concentrations, affects the distribution of cleave-site motifs, and increases the fraction of circulating fetal DNA during pregnancy. For the application of NIPT, and potentially other cfDNA-based tests, this variant has direct clinical consequences, as it increases the odds of inconclusive results and impairs the sensitivity of NIPT by causing predictors to overestimate the fetal fraction.

TWAS facilitates gene-scale trait genetic dissection through gene expression, structural variations, and alternative splicing in soybean

A genome-wide association study (GWAS) identifies trait-associated loci, but identifying the causal genes can be a bottleneck, due in part to slow decay of linkage disequilibrium (LD). A transcriptome-wide association study (TWAS) addresses this issue by identifying gene expression-phenotype associations or integrating gene expression quantitative trait loci with GWAS results. Here, we used self-pollinated soybean (Glycine max [L.] Merr.) as a model to evaluate the application of TWAS to the genetic dissection of traits in plant species with slow LD decay. We generated RNA sequencing data for a soybean diversity panel and identified the genetic expression regulation of 29 286 soybean genes. Different TWAS solutions were less affected by LD and were robust to the source of expression, identifing known genes related to traits from different tissues and developmental stages. The novel pod-color gene L2 was identified via TWAS and functionally validated by genome editing. By introducing a new exon proportion feature, we significantly improved the detection of expression variations that resulted from structural variations and alternative splicing. As a result, the genes identified through our TWAS approach exhibited a diverse range of causal variations, including SNPs, insertions or deletions, gene fusion, copy number variations, and alternative splicing. Using this approach, we identified genes associated with flowering time, including both previously known genes and novel genes that had not previously been linked to this trait, providing insights complementary to those from GWAS. In summary, this study supports the application of TWAS for candidate gene identification in species with low rates of LD decay.

Dissecting shared genetic architecture between depression and body mass index

BACKGROUND

A growing body of evidence supports the comorbidity between depression (DEP) and obesity, yet the genetic mechanisms underlying this association remain unclear. Our study explored the shared genetic architecture and causal associations of DEP with BMI.

METHODS

We investigated the multigene overlap and genetic correlation between DEP (N > 1.3 million) and BMI (N = 806,834) based on genome-wide association studies (GWAS) and using the bivariate causal mixture model and linkage disequilibrium score regression (LDSC). The causal association was explored by bi-directional Mendelian randomization (MR). Common risk loci were identified through cross-trait meta-analyses. Stratified LDSC and multi-marker gene annotation analyses were applied to investigate single-nucleotide polymorphisms enrichment across tissue types, cell types, and functional categories. Finally, we explored shared functional genes by Summary Data-Based Mendelian Randomization (SMR) and further detected differential expression genes (DEG) in brain tissues of individuals with depression and obesity.

RESULTS

We found a positive genetic correlation between DEP and BMI (rg = 0.19, P = 4.07 × 10-26), which was more evident in local genomic regions. Cross-trait meta-analyses identified 16 shared genetic loci, 5 of which were newly identified, and they had influence on both diseases in the same direction. MR analysis showed a bidirectional causal association between DEP and BMI, with comparable effect sizes estimated in both directions. Combined with gene expression information, we found that genetic correlations between DEP and BMI were enriched in 6 brain regions, predominantly in the nucleus accumbens and anterior cingulate cortex. Moreover, 6 specific cell types and 23 functional genes were found to have an impact on both DEP and BMI across the brain regions. Of which, NEGR1 was identified as the most significant functional gene and associated with DEP and BMI at the genome-wide significance level (P < 5 × 10-8). Compared with healthy controls, the expression levels of NEGR1 gene were significant lower in brain tissues of individuals with depression and obesity.

CONCLUSIONS

Our study reveals shared genetic basis underpinnings between DEP and BMI, including genetic correlations and common genes. These insights offer novel opportunities and avenues for future research into their comorbidities.

Linking candidate causal autoimmune variants to T cell networks using genetic and epigenetic screens in primary human T cells

Genetic variants associated with autoimmune diseases are highly enriched within putative cis -regulatory regions of CD4 + T cells, suggesting that they alter disease risk via changes in gene regulation. However, very few genetic variants have been shown to affect T cell gene expression or function. We tested >18,000 autoimmune disease-associated variants for allele-specific expression using massively parallel reporter assays in primary human CD4 + T cells. The 545 expression-modulating variants (emVars) identified greatly enrich for likely causal variants. We provide evidence that many emVars are mediated by common upstream regulatory conduits, and that putative target genes of primary T cell emVars are highly enriched within a lymphocyte activation network. Using bulk and single-cell CRISPR-interference screens, we confirm that emVar-containing T cell cis -regulatory elements modulate both known and novel target genes that regulate T cell proliferation, providing plausible mechanisms by which these variants alter autoimmune disease risk.

Genetic analyses of 104 phenotypes in 20,900 Chinese pregnant women reveal pregnancy-specific discoveries

Monitoring biochemical phenotypes during pregnancy is vital for maternal and fetal health, allowing early detection and management of pregnancy-related conditions to ensure safety for both. Here, we conducted a genetic analysis of 104 pregnancy phenotypes in 20,900 Chinese women. The genome-wide association study (GWAS) identified a total of 410 trait-locus associations, with 71.71% reported previously. Among the 116 novel hits for 45 phenotypes, 83 were successfully replicated. Among them, 31 were defined as potentially pregnancy-specific associations, including creatine and HELLPAR and neutrophils and ESR1, with subsequent analysis revealing enrichments in estrogen-related pathways and female reproductive tissues. The partitioning heritability underscored the significant roles of fetal blood, embryoid bodies, and female reproductive organs in pregnancy hematology and birth outcomes. Pathway analysis confirmed the intricate interplay of hormone and immune regulation, metabolism, and cell cycle during pregnancy. This study contributes to the understanding of genetic influences on pregnancy phenotypes and their implications for maternal health.

Genes with differential expression across ancestries are enriched in ancestry-specific disease effects likely due to gene-by-environment interactions

Multi-ancestry genome-wide association studies (GWASs) have highlighted the existence of variants with ancestry-specific effect sizes. Understanding where and why these ancestry-specific effects occur is fundamental to understanding the genetic basis of human diseases and complex traits. Here, we characterized genes differentially expressed across ancestries (ancDE genes) at the cell-type level by leveraging single-cell RNA-sequencing data in peripheral blood mononuclear cells for 21 individuals with East Asian (EAS) ancestry and 23 individuals with European (EUR) ancestry (172,385 cells); then, we tested whether variants surrounding those genes were enriched in disease variants with ancestry-specific effect sizes by leveraging ancestry-matched GWASs of 31 diseases and complex traits (average n ∼ 90,000 and ∼ 267,000 in EAS and EUR, respectively). We observed that ancDE genes tended to be cell-type specific and enriched in genes interacting with the environment and in variants with ancestry-specific disease effect sizes, which suggests cell-type-specific, gene-by-environment interactions shared between regulatory and disease architectures. Finally, we illustrated how different environments might have led to ancestry-specific myeloid cell leukemia 1 (MCL1) expression in B cells and ancestry-specific allele effect sizes in lymphocyte count GWASs for variants surrounding MCL1. Our results imply that large single-cell and GWAS datasets from diverse ancestries are required to improve our understanding of human diseases.

Comprehensive Evaluation of the Genetic Basis of Keratoconus: New Perspectives for Clinical Translation

Purpose

Keratoconus (KC) is a corneal disorder with complex etiology, apparently involving both genetic and environmental factors, characterized by progressive thinning and protrusion of the cornea. We aimed to identify novel genetic regions associated with KC susceptibility, elucidate relevant genes for disease development, and explore the translational implications for therapeutic intervention and risk assessment.

Methods

We conducted a genome-wide association study (GWAS) that integrated previously published data with newly generated genotyping data from an independent European cohort. To evaluate the clinical translation of our results, we performed functional annotation, gene prioritization, polygenic risk score (PRS), and drug repositioning analyses.

Results

We identified two novel genetic loci associated with KC, with rs2806689 and rs807037 emerging as lead variants (P = 1.71E-08, odds ratio [OR] = 0.88; P = 1.93E-08, OR = 1.16, respectively). Most importantly, we identified 315 candidate genes influenced by confirmed KC-associated variants. Among these, MINK1 was found to play a pivotal role in KC pathogenesis through the WNT signaling pathway. Moreover, we developed a PRS model that successfully differentiated KC patients from controls (P = 7.61E-16; area under the curve = 0.713). This model has the potential to identify individuals at high risk for developing KC, which could be instrumental in early diagnosis and management. Additionally, our drug repositioning analysis identified acetylcysteine as a potential treatment option for KC, opening up new avenues for therapeutic intervention.

Conclusions

Our study provides valuable insights into the genetic and molecular basis of KC, offering new targets for therapy and highlighting the clinical utility of PRS models in predicting disease risk.

A cell type-aware framework for nominating non-coding variants in Mendelian regulatory disorders

Unsolved Mendelian cases often lack obvious pathogenic coding variants, suggesting potential non-coding etiologies. Here, we present a single cell multi-omic framework integrating embryonic mouse chromatin accessibility, histone modification, and gene expression assays to discover cranial motor neuron (cMN) cis-regulatory elements and subsequently nominate candidate non-coding variants in the congenital cranial dysinnervation disorders (CCDDs), a set of Mendelian disorders altering cMN development. We generate single cell epigenomic profiles for ~86,000 cMNs and related cell types, identifying ~250,000 accessible regulatory elements with cognate gene predictions for ~145,000 putative enhancers. We evaluate enhancer activity for 59 elements using an in vivo transgenic assay and validate 44 (75%), demonstrating that single cell accessibility can be a strong predictor of enhancer activity. Applying our cMN atlas to 899 whole genome sequences from 270 genetically unsolved CCDD pedigrees, we achieve significant reduction in our variant search space and nominate candidate variants predicted to regulate known CCDD disease genes MAFB, PHOX2A, CHN1, and EBF3 - as well as candidates in recurrently mutated enhancers through peak- and gene-centric allelic aggregation. This work delivers non-coding variant discoveries of relevance to CCDDs and a generalizable framework for nominating non-coding variants of potentially high functional impact in other Mendelian disorders.

Investigating the Shared Genetic Architecture Between Leukocyte Telomere Length and Prostate Cancer

PURPOSE

Evidence of an association between leukocyte telomere length (LTL) and prostate cancer (PCa) is accumulating; however, their shared genetic basis remains unclear.

MATERIALS AND METHODS

Using summary statistics obtained from the genome-wide association study (GWAS), we quantified the global and local genetic correlations between two traits. Subsequently, we identified potential pleiotropic loci, common tissue-enriched regions, and risk gene loci while inferring assumed causal relationships.

RESULTS

Our study demonstrated a global genetic correlation between LTL and PCa (genetic correlation=0.066, p=0.017), which was further confirmed in local genomic regions. Cross-trait GWAS meta-analysis revealed 44 shared loci, including 10 novel pleiotropic single nucleotide polymorphisms appearing concurrently in significant local genetic correlation regions. Notably, two new loci (rs9419958; rs3730668) were additionally validated to co-localize. For the first time, we identified a significant shared genetic enrichment of both traits in the small intestine tissue at the terminal ileum, with functional genes in this region affecting both LTL and PCa. Concurrently, Mendelian randomization analysis indicated a positive causal relationship between LTL and PCa.

CONCLUSIONS

In conclusion, our study makes a significant contribution to the ongoing debate concerning the potential association between longer LTL and a higher risk of PCa. Additionally, we provide new evidence for the development of therapeutic targets for PCa and propose new directions for future risk prediction in this regard.

Expression patterns of housekeeping genes and tissue-specific genes in black goats across multiple tissues

Black goats are a significant meat breed in southern China. To investigate the expression patterns and biological functions of genes in various tissues of black goats, we analyzed housekeeping genes (HKGs), tissue-specific genes (TSGs), and hub genes (HUBGs) across 23 tissues. Additionally, we analyzed HKGs in 13 tissues under different feeding conditions. We identified 2968 HKGs, including six important ones. Interestingly, HKGs in grazing black goats demonstrated higher and more stable expression levels. We discovered a total of 9912 TSGs, including 134 newly identified ones. The number of TSGs for mRNA and lncRNA were nearly equal, with 127 mRNA TSGs expressed solely in one tissue. Additionally, the predicted functions of tissue-specific long non-coding RNAs (lncRNAs) targeting mRNAs corresponded with the physiological functions of the tissues.Weighted gene co-expression network analysis (WGCNA) constructed 30 modules, where the dark grey module consists almost entirely of HKGs, and TSGs are located in modules most correlated with their respective tissues. Additionally, we identified 289 HUBGs, which are involved in regulating the physiological functions of their respective tissues. Overall, these identified HKGs, TSGs, and HUBGs provide a foundation for studying the molecular mechanisms affecting the genetic and biological processes of complex traits in black goats.

Investigating the shared genetic basis of inflammatory bowel disease and systemic lupus erythematosus using genetic overlap analysis

BACKGROUND

Inflammatory bowel disease (IBD) and systemic lupus erythematosus (SLE) are autoimmune diseases that often coexist clinically. This phenomenon might be due to shared genetic components.

METHODS

Genome-wide association study (GWAS) data for IBD and SLE were analyzed to determine both global and local genetic correlations using three methodologies: linkage disequilibrium score regression (LDSC), genetic covariance analyzer (GNOVA), and SUPERGNOVA. The genetic overlap and risk loci were subsequently examined using the conditional/conjunctional false discovery rate (cond/conjFDR) statistical framework. Furthermore, a multi-trait analysis of MTAG was employed to validate the loci, followed by an LDSC analysis focusing on tissue-specific gene expression.

RESULTS

GWAS findings demonstrated a marked global genetic correlation between IBD (including Crohn's disease and ulcerative colitis) and SLE. Locally, SLE showed a strong association with IBD and Crohn's disease on chromosomes 10, 19, and 22. ConjFDR analysis confirmed the genetic overlap and identified relevant genetic risk loci. MTAG further validated several shared susceptibility genes. Additionally, the LDSC-SEG analysis results indicate that IBD (including CD and UC) and SLE are jointly enriched in the tissues of Spleen and Whole Blood.

CONCLUSION

This study confirms a genetic overlap between IBD and SLE, identifying marked comorbid genes and offering new insights for treating these diseases.

Rhinovirus infection of airway epithelial cells uncovers the non-ciliated subset as a likely driver of genetic risk to childhood-onset asthma

Asthma is a complex disease caused by genetic and environmental factors. Studies show that wheezing during rhinovirus infection correlates with childhood asthma development. Over 150 non-coding risk variants for asthma have been identified, many affecting gene regulation in T cells, but the effects of most risk variants remain unknown. We hypothesized that airway epithelial cells could also mediate genetic susceptibility to asthma given they are the first line of defense against respiratory viruses and allergens. We integrated genetic data with transcriptomics of airway epithelial cells subject to different stimuli. We demonstrate that rhinovirus infection significantly upregulates childhood-onset asthma-associated genes, particularly in non-ciliated cells. This enrichment is also observed with influenza infection but not with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) or cytokine activation. Overall, our results suggest that rhinovirus infection is an environmental factor that interacts with genetic risk factors through non-ciliated airway epithelial cells to drive childhood-onset asthma.

Genetic Implication of Specific Glutamatergic Neurons of the Prefrontal Cortex in the Pathophysiology of Schizophrenia

Background

The prefrontal cortex (PFC) has been strongly implicated in the pathophysiology of schizophrenia. Here, we combined high-resolution single-nuclei RNA sequencing data from the human PFC with large-scale genomic data for schizophrenia to identify constituent cell populations likely to mediate genetic liability to the disorder.

Methods

Gene expression specificity values were calculated from a single-nuclei RNA sequencing dataset comprising 84 cell populations from the human PFC, spanning gestation to adulthood. Enrichment of schizophrenia common variant liability and burden of rare protein-truncating coding variants were tested in genes with high expression specificity for each cell type. We also explored schizophrenia common variant associations in relation to gene expression across the developmental trajectory of implicated neurons.

Results

Common risk variation for schizophrenia was prominently enriched in genes with high expression specificity for a population of mature layer 4 glutamatergic neurons emerging in infancy. Common variant liability to schizophrenia increased along the developmental trajectory of this neuronal population. Fine-mapped genes at schizophrenia genome-wide association study risk loci had significantly higher expression specificity than other genes in these neurons and in a population of layer 5/6 glutamatergic neurons. People with schizophrenia had a higher rate of rare protein-truncating coding variants in genes expressed by cells of the PFC than control individuals, but no cell population was significantly enriched above this background rate.

Conclusions

We identified a population of layer 4 glutamatergic PFC neurons likely to be particularly affected by common variant genetic risk for schizophrenia, which may contribute to disturbances in thalamocortical connectivity in the condition.

Transcriptomic dysregulation and autistic-like behaviors in Kmt2c haploinsufficient mice rescued by an LSD1 inhibitor

Recent studies have consistently demonstrated that the regulation of chromatin and gene transcription plays a pivotal role in the pathogenesis of neurodevelopmental disorders. Among many genes involved in these pathways, KMT2C, encoding one of the six known histone H3 lysine 4 (H3K4) methyltransferases in humans and rodents, was identified as a gene whose heterozygous loss-of-function variants are causally associated with autism spectrum disorder (ASD) and the Kleefstra syndrome phenotypic spectrum. However, little is known about how KMT2C haploinsufficiency causes neurodevelopmental deficits and how these conditions can be treated. To address this, we developed and analyzed genetically engineered mice with a heterozygous frameshift mutation of Kmt2c (Kmt2c+/fs mice) as a disease model with high etiological validity. In a series of behavioral analyses, the mutant mice exhibit autistic-like behaviors such as impairments in sociality, flexibility, and working memory, demonstrating their face validity as an ASD model. To investigate the molecular basis of the observed abnormalities, we performed a transcriptomic analysis of their bulk adult brains and found that ASD risk genes were specifically enriched in the upregulated differentially expressed genes (DEGs), whereas KMT2C peaks detected by ChIP-seq were significantly co-localized with the downregulated genes, suggesting an important role of putative indirect effects of Kmt2c haploinsufficiency. We further performed single-cell RNA sequencing of newborn mouse brains to obtain cell type-resolved insights at an earlier stage. By integrating findings from ASD exome sequencing, genome-wide association, and postmortem brain studies to characterize DEGs in each cell cluster, we found strong ASD-associated transcriptomic changes in radial glia and immature neurons with no obvious bias toward upregulated or downregulated DEGs. On the other hand, there was no significant gross change in the cellular composition. Lastly, we explored potential therapeutic agents and demonstrate that vafidemstat, a lysine-specific histone demethylase 1 (LSD1) inhibitor that was effective in other models of neuropsychiatric/neurodevelopmental disorders, ameliorates impairments in sociality but not working memory in adult Kmt2c+/fs mice. Intriguingly, the administration of vafidemstat was shown to alter the vast majority of DEGs in the direction to normalize the transcriptomic abnormalities in the mutant mice (94.3 and 82.5% of the significant upregulated and downregulated DEGs, respectively, P < 2.2 × 10-16, binomial test), which could be the molecular mechanism underlying the behavioral rescuing. In summary, our study expands the repertoire of ASD models with high etiological and face validity, elucidates the cell-type resolved molecular alterations due to Kmt2c haploinsufficiency, and demonstrates the efficacy of an LSD1 inhibitor that might be generalizable to multiple categories of psychiatric disorders along with a better understanding of its presumed mechanisms of action.

Deciphering the impact of genomic variation on function

Our genomes influence nearly every aspect of human biology-from molecular and cellular functions to phenotypes in health and disease. Studying the differences in DNA sequence between individuals (genomic variation) could reveal previously unknown mechanisms of human biology, uncover the basis of genetic predispositions to diseases, and guide the development of new diagnostic tools and therapeutic agents. Yet, understanding how genomic variation alters genome function to influence phenotype has proved challenging. To unlock these insights, we need a systematic and comprehensive catalogue of genome function and the molecular and cellular effects of genomic variants. Towards this goal, the Impact of Genomic Variation on Function (IGVF) Consortium will combine approaches in single-cell mapping, genomic perturbations and predictive modelling to investigate the relationships among genomic variation, genome function and phenotypes. IGVF will create maps across hundreds of cell types and states describing how coding variants alter protein activity, how noncoding variants change the regulation of gene expression, and how such effects connect through gene-regulatory and protein-interaction networks. These experimental data, computational predictions and accompanying standards and pipelines will be integrated into an open resource that will catalyse community efforts to explore how our genomes influence biology and disease across populations.

Epidemiologic and genetic associations between primary biliary cholangitis and extrahepatic rheumatic diseases

Patients with primary biliary cholangitis (PBC) commonly experience extrahepatic rheumatic diseases. However, the epidemiologic and genetic associations as well as causal relationship between PBC and these extrahepatic conditions remain undetermined. In this study, we first conducted systematic review and meta-analyses by analyzing 73 studies comprising 334,963 participants across 17 countries and found strong phenotypic associations between PBC and rheumatic diseases. Next, we utilized large-scale genome-wide association study summary data to define the shared genetic architecture between PBC and rheumatic diseases, including rheumatoid arthritis (RA), systemic lupus erythematosus (SLE), systemic sclerosis (SSc) and Sjögren's syndrome (SS). We observed significant genetic correlations between PBC and each of the four rheumatic diseases. Pleiotropy and heritability enrichment analysis suggested the involvement of humoral immunity and interferon-associated processes for the comorbidity. Of note, we identified four variants shared between PBC and RA (rs80200208), SLE (rs9843053), and SSc (rs27524, rs3873182) using cross-trait meta-analysis. Additionally, several pleotropic loci for PBC and rheumatic diseases were found to share causal variants with gut microbes possessing immunoregulatory functions. Finally, Mendelian randomization revealed consistent evidence for a causal effect of PBC on RA, SLE, SSc, and SS, but no or inconsistent evidence for a causal effect of extrahepatic rheumatic diseases on PBC. Our study reveals a profound genetic overlap and causal relationships between PBC and extrahepatic rheumatic diseases, thus providing insights into shared biological mechanisms and novel therapeutic interventions.

Schizophrenia genomics: genetic complexity and functional insights

Determining the causes of schizophrenia has been a notoriously intractable problem, resistant to a multitude of investigative approaches over centuries. In recent decades, genomic studies have delivered hundreds of robust findings that implicate nearly 300 common genetic variants (via genome-wide association studies) and more than 20 rare variants (via whole-exome sequencing and copy number variant studies) as risk factors for schizophrenia. In parallel, functional genomic and neurobiological studies have provided exceptionally detailed information about the cellular composition of the brain and its interconnections in neurotypical individuals and, increasingly, in those with schizophrenia. Taken together, these results suggest unexpected complexity in the mechanisms that drive schizophrenia, pointing to the involvement of ensembles of genes (polygenicity) rather than single-gene causation. In this Review, we describe what we now know about the genetics of schizophrenia and consider the neurobiological implications of this information.

Genetic Implication of Prenatal GABAergic and Cholinergic Neuron Development in Susceptibility to Schizophrenia

BACKGROUND

The ganglionic eminences (GE) are fetal-specific structures that give rise to gamma-aminobutyric acid (GABA)- and acetylcholine-releasing neurons of the forebrain. Given the evidence for GABAergic, cholinergic, and neurodevelopmental disturbances in schizophrenia, we tested the potential involvement of GE neuron development in mediating genetic risk for the condition.

STUDY DESIGN

We combined data from a recent large-scale genome-wide association study of schizophrenia with single-cell RNA sequencing data from the human GE to test the enrichment of schizophrenia risk variation in genes with high expression specificity for developing GE cell populations. We additionally performed the single nuclei Assay for Transposase-Accessible Chromatin with Sequencing (snATAC-Seq) to map potential regulatory genomic regions operating in individual cell populations of the human GE, using these to test for enrichment of schizophrenia common genetic variant liability and to functionally annotate non-coding variants-associated with the disorder.

STUDY RESULTS

Schizophrenia common variant liability was enriched in genes with high expression specificity for developing neuron populations that are predicted to form dopamine D1 and D2 receptor-expressing GABAergic medium spiny neurons of the striatum, cortical somatostatin-positive GABAergic interneurons, calretinin-positive GABAergic neurons, and cholinergic neurons. Consistent with these findings, schizophrenia genetic risk was concentrated in predicted regulatory genomic sequence mapped in developing neuronal populations of the GE.

CONCLUSIONS

Our study implicates prenatal development of specific populations of GABAergic and cholinergic neurons in later susceptibility to schizophrenia, and provides a map of predicted regulatory genomic elements operating in cells of the GE.

Large-scale cross-trait genetic analysis highlights shared genetic backgrounds of autoimmune diseases

Disorders associated with the immune system burden multiple organs, although the shared biology exists across the diseases. Preceding family-based studies reveal that immune diseases are heritable to varying degrees, providing the basis for immunogenomics. The recent cost reduction in genetic analysis intensively promotes biobank-scale studies and the development of frameworks for statistical genetics. The accumulating multi-layer omics data, including genome-wide association studies (GWAS) and RNA-sequencing at single-cell resolution, enable us to dissect the genetic backgrounds of immune-related disorders. Although autoimmune and allergic diseases are generally categorized into different disease categories, epidemiological studies reveal the high incidence of autoimmune and allergic disease complications, suggesting the shared genetics and biology between the disease categories. Biobank resources and consortia cover multiple immune-related disorders to accumulate phenome-wide associations of genetic variants and enhance researchers to analyze the shared and heterogeneous genetic backgrounds. The emerging post-GWAS and integrative multi-omics analyses provide genetic and biological insights into the multicategorical disease associations.

Single-nucleus multiome analysis of human cerebellum in Alzheimer's disease-related dementia

Although human cerebellum is known to be neuropathologically impaired in Alzheimer's disease (AD) and AD-related dementias (ADRD), the cell type-specific transcriptional and epigenomic changes that contribute to this pathology are not well understood. Here, we report single-nucleus multiome (snRNA-seq and snATAC-seq) analysis of 103,861 nuclei isolated from cerebellum from 9 human cases of AD/ADRD and 8 controls, and with frontal cortex of 6 AD donors for additional comparison. Using peak-to-gene linkage analysis, we identified 431,834 significant linkages between gene expression and cell subtype-specific chromatin accessibility regions enriched for candidate cis-regulatory elements (cCREs). These cCREs were associated with AD/ADRD-specific transcriptomic changes and disease-related gene regulatory networks, especially for RAR Related Orphan Receptor A (RORA) and E74 Like ETS Transcription Factor 1 (ELF1) in cerebellar Purkinje cells and granule cells, respectively. Trajectory analysis of granule cell populations further identified disease-relevant transcription factors, such as RORA, and their regulatory targets. Finally, we prioritized two likely causal genes, including Seizure Related 6 Homolog Like 2 (SEZ6L2) in Purkinje cells and KAT8 Regulatory NSL Complex Subunit 1 (KANSL1) in granule cells, through integrative analysis of cCREs derived from snATAC-seq, genome-wide AD/ADRD loci, and Hi-C looping data. This first cell subtype-specific regulatory landscape in the human cerebellum identified here offer novel genomic and epigenomic insights into the neuropathology and pathobiology of AD/ADRD and other neurological disorders if broadly applied.

An integrative multi-context Mendelian randomization method for identifying risk genes across human tissues

Mendelian randomization (MR) provides valuable assessments of the causal effect of exposure on outcome, yet the application of conventional MR methods for mapping risk genes encounters new challenges. One of the issues is the limited availability of expression quantitative trait loci (eQTLs) as instrumental variables (IVs), hampering the estimation of sparse causal effects. Additionally, the often context- or tissue-specific eQTL effects challenge the MR assumption of consistent IV effects across eQTL and GWAS data. To address these challenges, we propose a multi-context multivariable integrative MR framework, mintMR, for mapping expression and molecular traits as joint exposures. It models the effects of molecular exposures across multiple tissues in each gene region, while simultaneously estimating across multiple gene regions. It uses eQTLs with consistent effects across more than one tissue type as IVs, improving IV consistency. A major innovation of mintMR involves employing multi-view learning methods to collectively model latent indicators of disease relevance across multiple tissues, molecular traits, and gene regions. The multi-view learning captures the major patterns of disease relevance and uses these patterns to update the estimated tissue relevance probabilities. The proposed mintMR iterates between performing a multi-tissue MR for each gene region and joint learning the disease-relevant tissue probabilities across gene regions, improving the estimation of sparse effects across genes. We apply mintMR to evaluate the causal effects of gene expression and DNA methylation for 35 complex traits using multi-tissue QTLs as IVs. The proposed mintMR controls genome-wide inflation and offers insights into disease mechanisms.

Principled distillation of UK Biobank phenotype data reveals underlying structure in human variation

Data within biobanks capture broad yet detailed indices of human variation, but biobank-wide insights can be difficult to extract due to complexity and scale. Here, using large-scale factor analysis, we distill hundreds of variables (diagnoses, assessments and survey items) into 35 latent constructs, using data from unrelated individuals with predominantly estimated European genetic ancestry in UK Biobank. These factors recapitulate known disease classifications, disentangle elements of socioeconomic status, highlight the relevance of psychiatric constructs to health and improve measurement of pro-health behaviours. We go on to demonstrate the power of this approach to clarify genetic signal, enhance discovery and identify associations between underlying phenotypic structure and health outcomes. In building a deeper understanding of ways in which constructs such as socioeconomic status, trauma, or physical activity are structured in the dataset, we emphasize the importance of considering the interwoven nature of the human phenome when evaluating public health patterns.

Regression convolutional neural network models implicate peripheral immune regulatory variants in the predisposition to Alzheimer's disease

Alzheimer's disease (AD) involves aggregation of amyloid β and tau, neuron loss, cognitive decline, and neuroinflammatory responses. Both resident microglia and peripheral immune cells have been associated with the immune component of AD. However, the relative contribution of resident and peripheral immune cell types to AD predisposition has not been thoroughly explored due to their similarity in gene expression and function. To study the effects of AD-associated variants on cis-regulatory elements, we train convolutional neural network (CNN) regression models that link genome sequence to cell type-specific levels of open chromatin, a proxy for regulatory element activity. We then use in silico mutagenesis of regulatory sequences to predict the relative impact of candidate variants across these cell types. We develop and apply criteria for evaluating our models and refine our models using massively parallel reporter assay (MPRA) data. Our models identify multiple AD-associated variants with a greater predicted impact in peripheral cells relative to microglia or neurons. Our results support their use as models to study the effects of AD-associated variants and even suggest that peripheral immune cells themselves may mediate a component of AD predisposition. We make our library of CNN models and predictions available as a resource for the community to study immune and neurological disorders.

Current genomic deep learning models display decreased performance in cell type-specific accessible regions

BACKGROUND

A number of deep learning models have been developed to predict epigenetic features such as chromatin accessibility from DNA sequence. Model evaluations commonly report performance genome-wide; however, cis regulatory elements (CREs), which play critical roles in gene regulation, make up only a small fraction of the genome. Furthermore, cell type-specific CREs contain a large proportion of complex disease heritability.

RESULTS

We evaluate genomic deep learning models in chromatin accessibility regions with varying degrees of cell type specificity. We assess two modeling directions in the field: general purpose models trained across thousands of outputs (cell types and epigenetic marks) and models tailored to specific tissues and tasks. We find that the accuracy of genomic deep learning models, including two state-of-the-art general purpose models-Enformer and Sei-varies across the genome and is reduced in cell type-specific accessible regions. Using accessibility models trained on cell types from specific tissues, we find that increasing model capacity to learn cell type-specific regulatory syntax-through single-task learning or high capacity multi-task models-can improve performance in cell type-specific accessible regions. We also observe that improving reference sequence predictions does not consistently improve variant effect predictions, indicating that novel strategies are needed to improve performance on variants.

CONCLUSIONS

Our results provide a new perspective on the performance of genomic deep learning models, showing that performance varies across the genome and is particularly reduced in cell type-specific accessible regions. We also identify strategies to maximize performance in cell type-specific accessible regions.

The Genetic Architecture of the Human Corpus Callosum and its Subregions

The corpus callosum (CC) is the largest set of white matter fibers connecting the two hemispheres of the brain. In humans, it is essential for coordinating sensorimotor responses, performing associative/executive functions, and representing information in multiple dimensions. Understanding which genetic variants underpin corpus callosum morphometry, and their shared influence on cortical structure and susceptibility to neuropsychiatric disorders, can provide molecular insights into the CC's role in mediating cortical development and its contribution to neuropsychiatric disease. To characterize the morphometry of the midsagittal corpus callosum, we developed a publicly available artificial intelligence based tool to extract, parcellate, and calculate its total and regional area and thickness. Using the UK Biobank (UKB) and the Adolescent Brain Cognitive Development study (ABCD), we extracted measures of midsagittal corpus callosum morphometry and performed a genome-wide association study (GWAS) meta-analysis of European participants (combined N = 46,685). We then examined evidence for generalization to the non-European participants of the UKB and ABCD cohorts (combined N = 7,040). Post-GWAS analyses implicate prenatal intracellular organization and cell growth patterns, and high heritability in regions of open chromatin, suggesting transcriptional activity regulation in early development. Results suggest programmed cell death mediated by the immune system drives the thinning of the posterior body and isthmus. Global and local genetic overlap, along with causal genetic liability, between the corpus callosum, cerebral cortex, and neuropsychiatric disorders such as attention-deficit/hyperactivity and bipolar disorders were identified. These results provide insight into variability of corpus callosum development, its genetic influence on the cerebral cortex, and biological mechanisms related to neuropsychiatric dysfunction.

European and African-specific plasma protein-QTL and metabolite-QTL analyses identify ancestry-specific T2D effector proteins and metabolites

Initially focused on the European population, multiple genome-wide association studies (GWAS) of complex diseases, such as type-2 diabetes (T2D), have now extended to other populations. However, to date, few ancestry-matched omics datasets have been generated or further integrated with the disease GWAS to nominate the key genes and/or molecular traits underlying the disease risk loci. In this study, we generated and integrated plasma proteomics and metabolomics with array-based genotype datasets of European (EUR) and African (AFR) ancestries to identify ancestry-specific muti-omics quantitative trait loci (QTLs). We further applied these QTLs to ancestry-stratified T2D risk to pinpoint key proteins and metabolites underlying the disease-associated genetic loci. We nominated five proteins and four metabolites in the European group and one protein and one metabolite in the African group to be part of the molecular pathways of T2D risk in an ancestry-stratified manner. Our study demonstrates the integration of genetic and omic studies of different ancestries can be used to identify distinct effector molecular traits underlying the same disease across diverse populations. Specifically, in the AFR proteomic findings on T2D, we prioritized the protein QSOX2; while in the AFR metabolomic findings, we pinpointed the metabolite GlcNAc sulfate conjugate of C21H34O2 steroid. Neither of these findings overlapped with the corresponding EUR results.

A scalable and robust variance components method reveals insights into the architecture of gene-environment interactions underlying complex traits

Understanding the contribution of gene-environment interactions (GxE) to complex trait variation can provide insights into disease mechanisms, explain sources of heritability, and improve genetic risk prediction. While large biobanks with genetic and deep phenotypic data hold promise for obtaining novel insights into GxE, our understanding of GxE architecture in complex traits remains limited. We introduce a method to estimate the proportion of trait variance explained by GxE (GxE heritability) and additive genetic effects (additive heritability) across the genome and within specific genomic annotations. We show that our method is accurate in simulations and computationally efficient for biobank-scale datasets. We applied our method to common array SNPs (MAF ≥1%), fifty quantitative traits, and four environmental variables (smoking, sex, age, and statin usage) in unrelated white British individuals in the UK Biobank. We found 68 trait-E pairs with significant genome-wide GxE heritability (p<0.05/200) with a ratio of GxE to additive heritability of ≈6.8% on average. Analyzing ≈8 million imputed SNPs (MAF ≥0.1%), we documented an approximate 28% increase in genome-wide GxE heritability compared to array SNPs. We partitioned GxE heritability across minor allele frequency (MAF) and local linkage disequilibrium (LD) values, revealing that, like additive allelic effects, GxE allelic effects tend to increase with decreasing MAF and LD. Analyzing GxE heritability near genes highly expressed in specific tissues, we find significant brain-specific enrichment for body mass index (BMI) and basal metabolic rate in the context of smoking and adipose-specific enrichment for waist-hip ratio (WHR) in the context of sex.

Current genomic deep learning models display decreased performance in cell type specific accessible regions

Background

A number of deep learning models have been developed to predict epigenetic features such as chromatin accessibility from DNA sequence. Model evaluations commonly report performance genome-wide; however, cis regulatory elements (CREs), which play critical roles in gene regulation, make up only a small fraction of the genome. Furthermore, cell type specific CREs contain a large proportion of complex disease heritability.

Results

We evaluate genomic deep learning models in chromatin accessibility regions with varying degrees of cell type specificity. We assess two modeling directions in the field: general purpose models trained across thousands of outputs (cell types and epigenetic marks), and models tailored to specific tissues and tasks. We find that the accuracy of genomic deep learning models, including two state-of-the-art general purpose models - Enformer and Sei - varies across the genome and is reduced in cell type specific accessible regions. Using accessibility models trained on cell types from specific tissues, we find that increasing model capacity to learn cell type specific regulatory syntax - through single-task learning or high capacity multi-task models - can improve performance in cell type specific accessible regions. We also observe that improving reference sequence predictions does not consistently improve variant effect predictions, indicating that novel strategies are needed to improve performance on variants.

Conclusions

Our results provide a new perspective on the performance of genomic deep learning models, showing that performance varies across the genome and is particularly reduced in cell type specific accessible regions. We also identify strategies to maximize performance in cell type specific accessible regions.

Bayesian estimation of gene constraint from an evolutionary model with gene features

Measures of selective constraint on genes have been used for many applications, including clinical interpretation of rare coding variants, disease gene discovery and studies of genome evolution. However, widely used metrics are severely underpowered at detecting constraints for the shortest ~25% of genes, potentially causing important pathogenic mutations to be overlooked. Here we developed a framework combining a population genetics model with machine learning on gene features to enable accurate inference of an interpretable constraint metric, shet. Our estimates outperform existing metrics for prioritizing genes important for cell essentiality, human disease and other phenotypes, especially for short genes. Our estimates of selective constraint should have wide utility for characterizing genes relevant to human disease. Finally, our inference framework, GeneBayes, provides a flexible platform that can improve the estimation of many gene-level properties, such as rare variant burden or gene expression differences.

Genome-wide association study of major anxiety disorders in 122,341 European-ancestry cases identifies 58 loci and highlights GABAergic signaling

The major anxiety disorders (ANX; including generalized anxiety disorder, panic disorder, and phobias) are highly prevalent, often onset early, persist throughout life, and cause substantial global disability. Although distinct in their clinical presentations, they likely represent differential expressions of a dysregulated threat-response system. Here we present a genome-wide association meta-analysis comprising 122,341 European ancestry ANX cases and 729,881 controls. We identified 58 independent genome-wide significant ANX risk variants and 66 genes with robust biological support. In an independent sample of 1,175,012 self-report ANX cases and 1,956,379 controls, 51 of the 58 associated variants were replicated. As predicted by twin studies, we found substantial genetic correlation between ANX and depression, neuroticism, and other internalizing phenotypes. Follow-up analyses demonstrated enrichment in all major brain regions and highlighted GABAergic signaling as one potential mechanism underlying ANX genetic risk. These results advance our understanding of the genetic architecture of ANX and prioritize genes for functional follow-up studies.

Abdominal aortic aneurysm and cardiometabolic traits share strong genetic susceptibility to lipid metabolism and inflammation

Abdominal aortic aneurysm has a high heritability and often co-occurs with other cardiometabolic disorders, suggesting shared genetic susceptibility. We investigate this commonality leveraging recent GWAS studies of abdominal aortic aneurysm and 32 cardiometabolic traits. We find significant genetic correlations between abdominal aortic aneurysm and 21 of the cardiometabolic traits investigated, including causal relationships with coronary artery disease, hypertension, lipid traits, and blood pressure. For each trait pair, we identify shared causal variants, genes, and pathways, revealing that cholesterol metabolism and inflammation are shared most prominently. Additionally, we show the tissue and cell type specificity in the shared signals, with strong enrichment across traits in the liver, arteries, adipose tissues, macrophages, adipocytes, and fibroblasts. Finally, we leverage drug-gene databases to identify several lipid-lowering drugs and antioxidants with high potential to treat abdominal aortic aneurysm with comorbidities. Our study provides insight into the shared genetic mechanism between abdominal aortic aneurysm and cardiometabolic traits, and identifies potential targets for pharmacological intervention.

Transcriptome-wide characterization of genetic perturbations

Single cell CRISPR screens such as Perturb-seq enable transcriptomic profiling of genetic perturbations at scale. However, the data produced by these screens are often noisy due to cost and technical constraints, limiting power to detect true effects with conventional differential expression analyses. Here, we introduce TRanscriptome-wide Analysis of Differential Expression (TRADE), a statistical framework which estimates the transcriptome-wide distribution of true differential expression effects from noisy gene-level measurements. Within TRADE, we derive multiple novel, interpretable statistical metrics, including the "transcriptome-wide impact", an estimator of the overall transcriptional effect of a perturbation which is stable across sampling depths. We analyze new and published large-scale Perturb-seq datasets to show that many true transcriptional effects are not statistically significant, but detectable in aggregate with TRADE. In a genome-scale Perturb-seq screen, we find that a typical gene perturbation affects an estimated 45 genes, whereas a typical essential gene perturbation affects over 500 genes. An advantage of our approach is its ability to compare the transcriptomic effects of genetic perturbations across contexts and dosages despite differences in power. We use this ability to identify perturbations with cell-type dependent effects and to find examples of perturbations where transcriptional responses are not only larger in magnitude, but also qualitatively different, as a function of dosage. Lastly, we expand our analysis to case/control comparison of gene expression for neuropsychiatric conditions, finding that transcriptomic effect correlations are greater than genetic correlations for these diagnoses. TRADE lays an analytic foundation for the systematic comparison of genetic perturbation atlases, as well as differential expression experiments more broadly.

Understanding the genetic complexity of puberty timing across the allele frequency spectrum

Pubertal timing varies considerably and is associated with later health outcomes. We performed multi-ancestry genetic analyses on ~800,000 women, identifying 1,080 signals for age at menarche. Collectively, these explained 11% of trait variance in an independent sample. Women at the top and bottom 1% of polygenic risk exhibited ~11 and ~14-fold higher risks of delayed and precocious puberty, respectively. We identified several genes harboring rare loss-of-function variants in ~200,000 women, including variants in ZNF483, which abolished the impact of polygenic risk. Variant-to-gene mapping approaches and mouse gonadotropin-releasing hormone neuron RNA sequencing implicated 665 genes, including an uncharacterized G-protein-coupled receptor, GPR83, which amplified the signaling of MC3R, a key nutritional sensor. Shared signals with menopause timing at genes involved in DNA damage response suggest that the ovarian reserve might signal centrally to trigger puberty. We also highlight body size-dependent and independent mechanisms that potentially link reproductive timing to later life disease.

Genetic drivers and cellular selection of female mosaic X chromosome loss

Mosaic loss of the X chromosome (mLOX) is the most common clonal somatic alteration in leukocytes of female individuals1,2, but little is known about its genetic determinants or phenotypic consequences. Here, to address this, we used data from 883,574 female participants across 8 biobanks; 12% of participants exhibited detectable mLOX in approximately 2% of leukocytes. Female participants with mLOX had an increased risk of myeloid and lymphoid leukaemias. Genetic analyses identified 56 common variants associated with mLOX, implicating genes with roles in chromosomal missegregation, cancer predisposition and autoimmune diseases. Exome-sequence analyses identified rare missense variants in FBXO10 that confer a twofold increased risk of mLOX. Only a small fraction of associations was shared with mosaic Y chromosome loss, suggesting that distinct biological processes drive formation and clonal expansion of sex chromosome missegregation. Allelic shift analyses identified X chromosome alleles that are preferentially retained in mLOX, demonstrating variation at many loci under cellular selection. A polygenic score including 44 allelic shift loci correctly inferred the retained X chromosomes in 80.7% of mLOX cases in the top decile. Our results support a model in which germline variants predispose female individuals to acquiring mLOX, with the allelic content of the X chromosome possibly shaping the magnitude of clonal expansion.

Genetic correlation between smoking behavior and gastroesophageal reflux disease: insights from integrative multi-omics data

BACKGROUND

Observational studies have preliminarily revealed an association between smoking and gastroesophageal reflux disease (GERD). However, little is known about the causal relationship and shared genetic architecture between the two. This study aims to explore their common genetic correlations by leveraging genome-wide association studies (GWAS) of smoking behavior-specifically, smoking initiation (SI), never smoking (NS), ever smoking (ES), cigarettes smoked per day (CPD), age of smoking initiation(ASI) and GERD.

METHODS

Firstly, we conducted global cross-trait genetic correlation analysis and heritability estimation from summary statistics (HESS) to explore the genetic correlation between smoking behavior and GERD. Then, a joint cross-trait meta-analysis was performed to identify shared "pleiotropic SNPs" between smoking behavior and GERD, followed by co-localization analysis. Additionally, multi-marker analyses using annotation (MAGMA) were employed to explore the degree of enrichment of single nucleotide polymorphism (SNP) heritability in specific tissues, and summary data-based Mendelian randomization (SMR) was further utilized to investigate potential functional genes. Finally, Mendelian randomization (MR) analysis was conducted to explore the causal relationship between the smoking behavior and GERD.

RESULTS

Consistent genetic correlations were observed through global and local genetic correlation analyses, wherein SI, ES, and CPD showed significantly positive genetic correlations with GERD, while NS and ASI showed significantly negative correlations. HESS analysis also identified multiple significantly associated loci between them. Furthermore, three novel "pleiotropic SNPs" (rs4382592, rs200968, rs1510719) were identified through cross-trait meta-analysis and co-localization analysis to exist between SI, NS, ES, ASI, and GERD, mapping the genes MED27, HIST1H2BO, MAML3 as new pleiotropic genes between SI, NS, ES, ASI, and GERD. Moreover, both smoking behavior and GERD were found to be co-enriched in multiple brain tissues, with GMPPB, RNF123, and RBM6 identified as potential functional genes co-enriched in Cerebellar Hemisphere, Cerebellum, Cortex/Nucleus accumbens in SI and GERD, and SUOX identified in Caudate nucleus, Cerebellum, Cortex in NS and GERD. Lastly, consistent causal relationships were found through MR analysis, indicating that SI, ES, and CPD increase the risk of GERD, while NS and higher ASI decrease the risk.

CONCLUSION

We identified genetic loci associated with smoking behavior and GERD, as well as brain tissue sites of shared enrichment, prioritizing three new pleiotropic genes and four new functional genes. Finally, the causal relationship between smoking behavior and GERD was demonstrated, providing insights for early prevention strategies for GERD.

Fine-mapping causal tissues and genes at disease-associated loci

Heritable diseases often manifest in a highly tissue-specific manner, with different disease loci mediated by genes in distinct tissues or cell types. We propose Tissue-Gene Fine-Mapping (TGFM), a fine-mapping method that infers the posterior probability (PIP) for each gene-tissue pair to mediate a disease locus by analyzing GWAS summary statistics (and in-sample LD) and leveraging eQTL data from diverse tissues to build cis-predicted expression models; TGFM also assigns PIPs to causal variants that are not mediated by gene expression in assayed genes and tissues. TGFM accounts for both co-regulation across genes and tissues and LD between SNPs (generalizing existing fine-mapping methods), and incorporates genome-wide estimates of each tissue's contribution to disease as tissue-level priors. TGFM was well-calibrated and moderately well-powered in simulations; unlike previous methods, TGFM was able to attain correct calibration by modeling uncertainty in cis-predicted expression models. We applied TGFM to 45 UK Biobank diseases/traits (average N = 316K) using eQTL data from 38 GTEx tissues. TGFM identified an average of 147 PIP > 0.5 causal genetic elements per disease/trait, of which 11% were gene-tissue pairs. Implicated gene-tissue pairs were concentrated in known disease-critical tissues, and causal genes were strongly enriched in disease-relevant gene sets. Causal gene-tissue pairs identified by TGFM recapitulated known biology (e.g., TPO-thyroid for Hypothyroidism), but also included biologically plausible novel findings (e.g., SLC20A2-artery aorta for Diastolic blood pressure). Further application of TGFM to single-cell eQTL data from 9 cell types in peripheral blood mononuclear cells (PBMC), analyzed jointly with GTEx tissues, identified 30 additional causal gene-PBMC cell type pairs at PIP > 0.5-primarily for autoimmune disease and blood cell traits, including the biologically plausible example of CD52 in classical monocyte cells for Monocyte count. In conclusion, TGFM is a robust and powerful method for fine-mapping causal tissues and genes at disease-associated loci.

Revealing brain cell-stratified causality through dissecting causal variants according to their cell-type-specific effects on gene expression

The human brain has been implicated in the pathogenesis of several complex diseases. Taking advantage of single-cell techniques, genome-wide association studies (GWAS) have taken it a step further and revealed brain cell-type-specific functions for disease loci. However, genetic causal associations inferred by Mendelian randomization (MR) studies usually include all instrumental variables from GWAS, which hampers the understanding of cell-specific causality. Here, we developed an analytical framework, Cell-Stratified MR (csMR), to investigate cell-stratified causality through colocalizing GWAS signals with single-cell eQTL from different brain cells. By applying to obesity-related traits, our results demonstrate the cell-type-specific effects of GWAS variants on gene expression, and indicate the benefits of csMR to identify cell-type-specific causal effect that is often hidden from bulk analyses. We also found csMR valuable to reveal distinct causal pathways between different obesity indicators. These findings suggest the value of our approach to prioritize target cells for extending genetic causation studies.

An integrative framework to prioritize genes in more than 500 loci associated with body mass index

Obesity is a major risk factor for a myriad of diseases, affecting >600 million people worldwide. Genome-wide association studies (GWASs) have identified hundreds of genetic variants that influence body mass index (BMI), a commonly used metric to assess obesity risk. Most variants are non-coding and likely act through regulating genes nearby. Here, we apply multiple computational methods to prioritize the likely causal gene(s) within each of the 536 previously reported GWAS-identified BMI-associated loci. We performed summary-data-based Mendelian randomization (SMR), FINEMAP, DEPICT, MAGMA, transcriptome-wide association studies (TWASs), mutation significance cutoff (MSC), polygenic priority score (PoPS), and the nearest gene strategy. Results of each method were weighted based on their success in identifying genes known to be implicated in obesity, ranking all prioritized genes according to a confidence score (minimum: 0; max: 28). We identified 292 high-scoring genes (≥11) in 264 loci, including genes known to play a role in body weight regulation (e.g., DGKI, ANKRD26, MC4R, LEPR, BDNF, GIPR, AKT3, KAT8, MTOR) and genes related to comorbidities (e.g., FGFR1, ISL1, TFAP2B, PARK2, TCF7L2, GSK3B). For most of the high-scoring genes, however, we found limited or no evidence for a role in obesity, including the top-scoring gene BPTF. Many of the top-scoring genes seem to act through a neuronal regulation of body weight, whereas others affect peripheral pathways, including circadian rhythm, insulin secretion, and glucose and carbohydrate homeostasis. The characterization of these likely causal genes can increase our understanding of the underlying biology and offer avenues to develop therapeutics for weight loss.

Genome-wide association studies and Mendelian randomization analyses provide insights into the causes of early-onset colorectal cancer

BACKGROUND

The incidence of early-onset colorectal cancer (EOCRC; diagnosed <50 years of age) is rising globally; however, the causes underlying this trend are largely unknown. CRC has strong genetic and environmental determinants, yet common genetic variants and causal modifiable risk factors underlying EOCRC are unknown. We conducted the first EOCRC-specific genome-wide association study (GWAS) and Mendelian randomization (MR) analyses to explore germline genetic and causal modifiable risk factors associated with EOCRC.

PATIENTS AND METHODS

We conducted a GWAS meta-analysis of 6176 EOCRC cases and 65 829 controls from the Genetics and Epidemiology of Colorectal Cancer Consortium (GECCO), the Colorectal Transdisciplinary Study (CORECT), the Colon Cancer Family Registry (CCFR), and the UK Biobank. We then used the EOCRC GWAS to investigate 28 modifiable risk factors using two-sample MR.

RESULTS

We found two novel risk loci for EOCRC at 1p34.1 and 4p15.33, which were not previously associated with CRC risk. We identified a deleterious coding variant (rs36053993, G396D) at polyposis-associated DNA repair gene MUTYH (odds ratio 1.80, 95% confidence interval 1.47-2.22) but show that most of the common genetic susceptibility was from noncoding signals enriched in epigenetic markers present in gastrointestinal tract cells. We identified new EOCRC-susceptibility genes, and in addition to pathways such as transforming growth factor (TGF) β, suppressor of Mothers Against Decapentaplegic (SMAD), bone morphogenetic protein (BMP) and phosphatidylinositol kinase (PI3K) signaling, our study highlights a role for insulin signaling and immune/infection-related pathways in EOCRC. In our MR analyses, we found novel evidence of probable causal associations for higher levels of body size and metabolic factors-such as body fat percentage, waist circumference, waist-to-hip ratio, basal metabolic rate, and fasting insulin-higher alcohol drinking, and lower education attainment with increased EOCRC risk.

CONCLUSIONS

Our novel findings indicate inherited susceptibility to EOCRC and suggest modifiable lifestyle and metabolic targets that could also be used to risk-stratify individuals for personalized screening strategies or other interventions.

Multi-ancestry meta-analysis of tobacco use disorder identifies 461 potential risk genes and reveals associations with multiple health outcomes

Tobacco use disorder (TUD) is the most prevalent substance use disorder in the world. Genetic factors influence smoking behaviours and although strides have been made using genome-wide association studies to identify risk variants, most variants identified have been for nicotine consumption, rather than TUD. Here we leveraged four US biobanks to perform a multi-ancestral meta-analysis of TUD (derived via electronic health records) in 653,790 individuals (495,005 European, 114,420 African American and 44,365 Latin American) and data from UK Biobank (ncombined = 898,680). We identified 88 independent risk loci; integration with functional genomic tools uncovered 461 potential risk genes, primarily expressed in the brain. TUD was genetically correlated with smoking and psychiatric traits from traditionally ascertained cohorts, externalizing behaviours in children and hundreds of medical outcomes, including HIV infection, heart disease and pain. This work furthers our biological understanding of TUD and establishes electronic health records as a source of phenotypic information for studying the genetics of TUD.