Estimating the selective effects of heterozygous protein-truncating variants from human exome data

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.

A deep catalogue of protein-coding variation in 983,578 individuals

Rare coding variants that substantially affect function provide insights into the biology of a gene1-3. However, ascertaining the frequency of such variants requires large sample sizes4-8. Here we present a catalogue of human protein-coding variation, derived from exome sequencing of 983,578 individuals across diverse populations. In total, 23% of the Regeneron Genetics Center Million Exome (RGC-ME) data come from individuals of African, East Asian, Indigenous American, Middle Eastern and South Asian ancestry. The catalogue includes more than 10.4 million missense and 1.1 million predicted loss-of-function (pLOF) variants. We identify individuals with rare biallelic pLOF variants in 4,848 genes, 1,751 of which have not been previously reported. From precise quantitative estimates of selection against heterozygous loss of function (LOF), we identify 3,988 LOF-intolerant genes, including 86 that were previously assessed as tolerant and 1,153 that lack established disease annotation. We also define regions of missense depletion at high resolution. Notably, 1,482 genes have regions that are depleted of missense variants despite being tolerant of pLOF variants. Finally, we estimate that 3% of individuals have a clinically actionable genetic variant, and that 11,773 variants reported in ClinVar with unknown significance are likely to be deleterious cryptic splice sites. To facilitate variant interpretation and genetics-informed precision medicine, we make this resource of coding variation from the RGC-ME dataset publicly accessible through a variant allele frequency browser.

Joint genotypic and phenotypic outcome modeling improves base editing variant effect quantification

CRISPR base editing screens enable analysis of disease-associated variants at scale; however, variable efficiency and precision confounds the assessment of variant-induced phenotypes. Here, we provide an integrated experimental and computational pipeline that improves estimation of variant effects in base editing screens. We use a reporter construct to measure guide RNA (gRNA) editing outcomes alongside their phenotypic consequences and introduce base editor screen analysis with activity normalization (BEAN), a Bayesian network that uses per-guide editing outcomes provided by the reporter and target site chromatin accessibility to estimate variant impacts. BEAN outperforms existing tools in variant effect quantification. We use BEAN to pinpoint common regulatory variants that alter low-density lipoprotein (LDL) uptake, implicating previously unreported genes. Additionally, through saturation base editing of LDLR, we accurately quantify missense variant pathogenicity that is consistent with measurements in UK Biobank patients and identify underlying structural mechanisms. This work provides a widely applicable approach to improve the power of base editing screens for disease-associated variant characterization.

Essentiality, protein-protein interactions and evolutionary properties are key predictors for identifying cancer-associated genes using machine learning

The distinctive nature of cancer as a disease prompts an exploration of the special characteristics the genes implicated in cancer exhibit. The identification of cancer-associated genes and their characteristics is crucial to further our understanding of this disease and enhanced likelihood of therapeutic drug targets success. However, the rate at which cancer genes are being identified experimentally is slow. Applying predictive analysis techniques, through the building of accurate machine learning models, is potentially a useful approach in enhancing the identification rate of these genes and their characteristics. Here, we investigated gene essentiality scores and found that they tend to be higher for cancer-associated genes compared to other protein-coding human genes. We built a dataset of extended gene properties linked to essentiality and used it to train a machine-learning model; this model reached 89% accuracy and > 0.85 for the Area Under Curve (AUC). The model showed that essentiality, evolutionary-related properties, and properties arising from protein-protein interaction networks are particularly effective in predicting cancer-associated genes. We were able to use the model to identify potential candidate genes that have not been previously linked to cancer. Prioritising genes that score highly by our methods could aid scientists in their cancer genes research.

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 constraint for the shortest ∼25% of genes, potentially causing important pathogenic mutations to be overlooked. 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 new 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 estimation of many gene-level properties, such as rare variant burden or gene expression differences.

Joint, multifaceted genomic analysis enables diagnosis of diverse, ultra-rare monogenic presentations

Genomics for rare disease diagnosis has advanced at a rapid pace due to our ability to perform "N-of-1" analyses on individual patients. The increasing sizes of ultra-rare, "N-of-1" disease cohorts internationally newly enables cohort-wide analyses for new discoveries, but well-calibrated statistical genetics approaches for jointly analyzing these patients are still under development.1,2 The Undiagnosed Diseases Network (UDN) brings multiple clinical, research and experimental centers under the same umbrella across the United States to facilitate and scale N-of-1 analyses. Here, we present the first joint analysis of whole genome sequencing data of UDN patients across the network. We apply existing and introduce new, well-calibrated statistical methods for prioritizing disease genes with de novo recurrence and compound heterozygosity. We also detect pathways enriched with candidate and known diagnostic genes. Our computational analysis, coupled with a systematic clinical review, recapitulated known diagnoses and revealed new disease associations. We make our gene-level findings and variant-level information across the cohort available in a public-facing browser (https://dbmi-bgm.github.io/udn-browser/). These results show that N-of-1 efforts should be supplemented by a joint genomic analysis across cohorts.

A foundational atlas of autism protein interactions reveals molecular convergence

Translating high-confidence (hc) autism spectrum disorder (ASD) genes into viable treatment targets remains elusive. We constructed a foundational protein-protein interaction (PPI) network in HEK293T cells involving 100 hcASD risk genes, revealing over 1,800 PPIs (87% novel). Interactors, expressed in the human brain and enriched for ASD but not schizophrenia genetic risk, converged on protein complexes involved in neurogenesis, tubulin biology, transcriptional regulation, and chromatin modification. A PPI map of 54 patient-derived missense variants identified differential physical interactions, and we leveraged AlphaFold-Multimer predictions to prioritize direct PPIs and specific variants for interrogation in Xenopus tropicalis and human forebrain organoids. A mutation in the transcription factor FOXP1 led to reconfiguration of DNA binding sites and altered development of deep cortical layer neurons in forebrain organoids. This work offers new insights into molecular mechanisms underlying ASD and describes a powerful platform to develop and test therapeutic strategies for many genetically-defined conditions.

Aggregation of recount3 RNA-seq data improves inference of consensus and tissue-specific gene co-expression networks

Background

Gene co-expression networks (GCNs) describe relationships among expressed genes key to maintaining cellular identity and homeostasis. However, the small sample size of typical RNA-seq experiments which is several orders of magnitude fewer than the number of genes is too low to infer GCNs reliably. recount3, a publicly available dataset comprised of 316,443 uniformly processed human RNA-seq samples, provides an opportunity to improve power for accurate network reconstruction and obtain biological insight from the resulting networks.

Results

We compared alternate aggregation strategies to identify an optimal workflow for GCN inference by data aggregation and inferred three consensus networks: a universal network, a non-cancer network, and a cancer network in addition to 27 tissue context-specific networks. Central network genes from our consensus networks were enriched for evolutionarily constrained genes and ubiquitous biological pathways, whereas central context-specific network genes included tissue-specific transcription factors and factorization based on the hubs led to clustering of related tissue contexts. We discovered that annotations corresponding to context-specific networks inferred from aggregated data were enriched for trait heritability beyond known functional genomic annotations and were significantly more enriched when we aggregated over a larger number of samples.

Conclusion

This study outlines best practices for network GCN inference and evaluation by data aggregation. We recommend estimating and regressing confounders in each data set before aggregation and prioritizing large sample size studies for GCN reconstruction. Increased statistical power in inferring context-specific networks enabled the derivation of variant annotations that were enriched for concordant trait heritability independent of functional genomic annotations that are context-agnostic. While we observed strictly increasing held-out log-likelihood with data aggregation, we noted diminishing marginal improvements. Future directions aimed at alternate methods for estimating confounders and integrating orthogonal information from modalities such as Hi-C and ChIP-seq can further improve GCN inference.

Unveiling recent and ongoing adaptive selection in human populations

Genome-wide scans for signals of selection have become a routine part of the analysis of population genomic variation datasets and have resulted in compelling evidence of selection during recent human evolution. This Essay spotlights methodological innovations that have enabled the detection of selection over very recent timescales, even in contemporary human populations. By harnessing large-scale genomic and phenotypic datasets, these new methods use different strategies to uncover connections between genotype, phenotype, and fitness. This Essay outlines the rationale and key findings of each strategy, discusses challenges in interpretation, and describes opportunities to improve detection and understanding of ongoing selection in human populations.

A probabilistic graphical model for estimating selection coefficient of missense variants from human population sequence data

Accurately predicting the effect of missense variants is a central problem in interpretation of genomic variation. Commonly used computational methods does not capture the quantitative impact on fitness in populations. We developed MisFit to estimate missense fitness effect using biobank-scale human population genome data. MisFit jointly models the effect at molecular level ( d ) and population level (selection coefficient, s ), assuming that in the same gene, missense variants with similar d have similar s . MisFit is a probabilistic graphical model that integrates deep neural network components and population genetics models efficiently with inductive bias based on biological causality of variant effect. We trained it by maximizing probability of observed allele counts in 236,017 European individuals. We show that s is informative in predicting frequency across ancestries and consistent with the fraction of de novo mutations given s . Finally, MisFit outperforms previous methods in prioritizing missense variants in individuals with neurodevelopmental disorders.

Using Computational Simulations to Model Deleterious Variation and Genetic Load in Natural Populations

AbstractDeleterious genetic variation is abundant in wild populations, and understanding the ecological and conservation implications of such variation is an area of active research. Genomic methods are increasingly used to quantify the impacts of deleterious variation in natural populations; however, these approaches remain limited by an inability to accurately predict the selective and dominance effects of mutations. Computational simulations of deleterious variation offer a complementary tool that can help overcome these limitations, although such approaches have yet to be widely employed. In this perspective article, we aim to encourage ecological and conservation genomics researchers to adopt greater use of computational simulations to aid in deepening our understanding of deleterious variation in natural populations. We first provide an overview of the components of a simulation of deleterious variation, describing the key parameters involved in such models. Next, we discuss several approaches for validating simulation models. Finally, we compare and validate several recently proposed deleterious mutation models, demonstrating that models based on estimates of selection parameters from experimental systems are biased toward highly deleterious mutations. We describe a new model that is supported by multiple orthogonal lines of evidence and provide example scripts for implementing this model (https://github.com/ckyriazis/simulations_review).

A mutation rate model at the basepair resolution identifies the mutagenic effect of polymerase III transcription

De novo mutations occur at substantially different rates depending on genomic location, sequence context and DNA strand. The success of methods to estimate selection intensity, infer demographic history and map rare disease genes, depends strongly on assumptions about the local mutation rate. Here we present Roulette, a genome-wide mutation rate model at basepair resolution that incorporates known determinants of local mutation rate. Roulette is shown to be more accurate than existing models. We use Roulette to refine the estimates of population growth within Europe by incorporating the full range of human mutation rates. The analysis of significant deviations from the model predictions revealed a tenfold increase in mutation rate in nearly all genes transcribed by polymerase III (Pol III), suggesting a new mutagenic mechanism. We also detected an elevated mutation rate within transcription factor binding sites restricted to sites actively used in testis and residing in promoters.

A deep catalog of protein-coding variation in 985,830 individuals

Coding variants that have significant impact on function can provide insights into the biology of a gene but are typically rare in the population. Identifying and ascertaining the frequency of such rare variants requires very large sample sizes. Here, we present the largest catalog of human protein-coding variation to date, derived from exome sequencing of 985,830 individuals of diverse ancestry to serve as a rich resource for studying rare coding variants. Individuals of African, Admixed American, East Asian, Middle Eastern, and South Asian ancestry account for 20% of this Exome dataset. Our catalog of variants includes approximately 10.5 million missense (54% novel) and 1.1 million predicted loss-of-function (pLOF) variants (65% novel, 53% observed only once). We identified individuals with rare homozygous pLOF variants in 4,874 genes, and for 1,838 of these this work is the first to document at least one pLOF homozygote. Additional insights from the RGC-ME dataset include 1) improved estimates of selection against heterozygous loss-of-function and identification of 3,459 genes intolerant to loss-of-function, 83 of which were previously assessed as tolerant to loss-of-function and 1,241 that lack disease annotations; 2) identification of regions depleted of missense variation in 457 genes that are tolerant to loss-of-function; 3) functional interpretation for 10,708 variants of unknown or conflicting significance reported in ClinVar as cryptic splice sites using splicing score thresholds based on empirical variant deleteriousness scores derived from RGC-ME; and 4) an observation that approximately 3% of sequenced individuals carry a clinically actionable genetic variant in the ACMG SF 3.1 list of genes. We make this important resource of coding variation available to the public through a variant allele frequency browser. We anticipate that this report and the RGC-ME dataset will serve as a valuable reference for understanding rare coding variation and help advance precision medicine efforts.

Scaling the discrete-time Wright-Fisher model to biobank-scale datasets

The discrete-time Wright-Fisher (DTWF) model and its diffusion limit are central to population genetics. These models can describe the forward-in-time evolution of allele frequencies in a population resulting from genetic drift, mutation, and selection. Computing likelihoods under the diffusion process is feasible, but the diffusion approximation breaks down for large samples or in the presence of strong selection. Existing methods for computing likelihoods under the DTWF model do not scale to current exome sequencing sample sizes in the hundreds of thousands. Here, we present a scalable algorithm that approximates the DTWF model with provably bounded error. Our approach relies on two key observations about the DTWF model. The first is that transition probabilities under the model are approximately sparse. The second is that transition distributions for similar starting allele frequencies are extremely close as distributions. Together, these observations enable approximate matrix-vector multiplication in linear (as opposed to the usual quadratic) time. We prove similar properties for Hypergeometric distributions, enabling fast computation of likelihoods for subsamples of the population. We show theoretically and in practice that this approximation is highly accurate and can scale to population sizes in the tens of millions, paving the way for rigorous biobank-scale inference. Finally, we use our results to estimate the impact of larger samples on estimating selection coefficients for loss-of-function variants. We find that increasing sample sizes beyond existing large exome sequencing cohorts will provide essentially no additional information except for genes with the most extreme fitness effects.

Large-scale causal discovery using interventional data sheds light on the regulatory network architecture of blood traits

Inference of directed biological networks is an important but notoriously challenging problem. We introduce inverse sparse regression (inspre), an approach to learning causal networks that leverages large-scale intervention-response data. Applied to 788 genes from the genome-wide perturb-seq dataset, inspre helps elucidate the network architecture of blood traits.

Ohnologs and SSD Paralogs Differ in Genomic and Expression Features Related to Dosage Constraints

Gene duplication is recognized as a critical process in genome evolution; however, many questions about this process remain unanswered. Although gene duplicability has been observed to differ by duplication mechanism and evolutionary rate, there is so far no broad characterization of its determinants. Many features correlate with this difference in duplicability; however, our ability to exploit these observations to advance our understanding of the role of duplication in evolution is hampered by limitations within existing work. In particular, the existence of methodological differences across studies impedes meaningful comparison. Here, we use consistent definitions of duplicability in the human lineage to explore these associations, allow resolution of the impact of confounding factors, and define the overall relevance of individual features. Using a classifier approach and controlling for the confounding effect of duplicate longevity, we find a subset of gene features important in differentiating genes duplicable by small-scale duplication from those duplicable by whole-genome duplication, revealing critical roles for gene dosage and expression costs in duplicability. We further delve into patterns of functional enrichment and find a lack of constraint on duplicate retention in any context for genes duplicable by small-scale duplication.

Deep multiple-instance learning accurately predicts gene haploinsufficiency and deletion pathogenicity

Copy number losses (deletions) are a major contributor to the etiology of severe genetic disorders. Although haploinsufficient genes play a critical role in deletion pathogenicity, current methods for deletion pathogenicity prediction fail to integrate multiple lines of evidence for haploinsufficiency at the gene level, limiting their power to pinpoint deleterious deletions associated with genetic disorders. Here we introduce DosaCNV, a deep multiple-instance learning framework that, for the first time, models deletion pathogenicity jointly with gene haploinsufficiency. By integrating over 30 gene-level features potentially predictive of haploinsufficiency, DosaCNV shows unmatched performance in prioritizing pathogenic deletions associated with a broad spectrum of genetic disorders. Furthermore, DosaCNV outperforms existing methods in predicting gene haploinsufficiency even though it is not trained on known haploinsufficient genes. Finally, DosaCNV leverages a state-of-the-art technique to quantify the contributions of individual gene-level features to haploinsufficiency, allowing for human-understandable explanations of model predictions. Altogether, DosaCNV is a powerful computational tool for both fundamental and translational research.

Repurposing Normal Chromosomal Microarray Data to Harbor Genetic Insights into Congenital Heart Disease

About 15% of congenital heart disease (CHD) patients have a known pathogenic copy number variant. The majority of their chromosomal microarray (CMA) tests are deemed normal. Diagnostic interpretation typically ignores microdeletions smaller than 100 kb. We hypothesized that unreported microdeletions are enriched for CHD genes. We analyzed "normal" CMAs of 1762 patients who were evaluated at a pediatric referral center, of which 319 (18%) had CHD. Using CMAs from monozygotic twins or replicates from the same individual, we established a size threshold based on probe count for the reproducible detection of small microdeletions. Genes in the microdeletions were sequentially filtered by their nominal association with a CHD diagnosis, the expression level in the fetal heart, and the deleteriousness of a loss-of-function mutation. The subsequent enrichment for CHD genes was assessed using the presence of known or potentially novel genes implicated by a large whole-exome sequencing study of CHD. The unreported microdeletions were modestly enriched for both known CHD genes and those of unknown significance identified using their de novo mutation in CHD patients. Our results show that readily available "normal" CMA data can be a fruitful resource for genetic discovery and that smaller deletions should receive more attention in clinical evaluation.

An unsupervised deep learning framework for predicting human essential genes from population and functional genomic data

BACKGROUND

The ability to accurately predict essential genes intolerant to loss-of-function (LOF) mutations can dramatically improve the identification of disease-associated genes. Recently, there have been numerous computational methods developed to predict human essential genes from population genomic data. While the existing methods are highly predictive of essential genes of long length, they have limited power in pinpointing short essential genes due to the sparsity of polymorphisms in the human genome.

RESULTS

Motivated by the premise that population and functional genomic data may provide complementary evidence for gene essentiality, here we present an evolution-based deep learning model, DeepLOF, to predict essential genes in an unsupervised manner. Unlike previous population genetic methods, DeepLOF utilizes a novel deep learning framework to integrate both population and functional genomic data, allowing us to pinpoint short essential genes that can hardly be predicted from population genomic data alone. Compared with previous methods, DeepLOF shows unmatched performance in predicting ClinGen haploinsufficient genes, mouse essential genes, and essential genes in human cell lines. Notably, at a false positive rate of 5%, DeepLOF detects 50% more ClinGen haploinsufficient genes than previous methods. Furthermore, DeepLOF discovers 109 novel essential genes that are too short to be identified by previous methods.

CONCLUSION

The predictive power of DeepLOF shows that it is a compelling computational method to aid in the discovery of essential genes.

Joint genotypic and phenotypic outcome modeling improves base editing variant effect quantification

CRISPR base editing screens are powerful tools for studying disease-associated variants at scale. However, the efficiency and precision of base editing perturbations vary, confounding the assessment of variant-induced phenotypic effects. Here, we provide an integrated pipeline that improves the estimation of variant impact in base editing screens. We perform high-throughput ABE8e-SpRY base editing screens with an integrated reporter construct to measure the editing efficiency and outcomes of each gRNA alongside their phenotypic consequences. We introduce BEAN, a Bayesian network that accounts for per-guide editing outcomes and target site chromatin accessibility to estimate variant impacts. We show this pipeline attains superior performance compared to existing tools in variant classification and effect size quantification. We use BEAN to pinpoint common variants that alter LDL uptake, implicating novel genes. Additionally, through saturation base editing of LDLR, we enable accurate quantitative prediction of the effects of missense variants on LDL-C levels, which aligns with measurements in UK Biobank individuals, and identify structural mechanisms underlying variant pathogenicity. This work provides a widely applicable approach to improve the power of base editor screens for disease-associated variant characterization.

DECIPHER: Improving Genetic Diagnosis Through Dynamic Integration of Genomic and Clinical Data

DECIPHER (Database of Genomic Variation and Phenotype in Humans Using Ensembl Resources) shares candidate diagnostic variants and phenotypic data from patients with genetic disorders to facilitate research and improve the diagnosis, management, and therapy of rare diseases. The platform sits at the boundary between genomic research and the clinical community. DECIPHER aims to ensure that the most up-to-date data are made rapidly available within its interpretation interfaces to improve clinical care. Newly integrated cardiac case-control data that provide evidence of gene-disease associations and inform variant interpretation exemplify this mission. New research resources are presented in a format optimized for use by a broad range of professionals supporting the delivery of genomic medicine. The interfaces within DECIPHER integrate and contextualize variant and phenotypic data, helping to determine a robust clinico-molecular diagnosis for rare-disease patients, which combines both variant classification and clinical fit. DECIPHER supports discovery research, connecting individuals within the rare-disease community to pursue hypothesis-driven research.

Partial gene suppression improves identification of cancer vulnerabilities when CRISPR-Cas9 knockout is pan-lethal

BACKGROUND

Hundreds of functional genomic screens have been performed across a diverse set of cancer contexts, as part of efforts such as the Cancer Dependency Map, to identify gene dependencies-genes whose loss of function reduces cell viability or fitness. Recently, large-scale screening efforts have shifted from RNAi to CRISPR-Cas9, due to superior efficacy and specificity. However, many effective oncology drugs only partially inhibit their protein targets, leading us to question whether partial suppression of genes using RNAi could reveal cancer vulnerabilities that are missed by complete knockout using CRISPR-Cas9. Here, we compare CRISPR-Cas9 and RNAi dependency profiles of genes across approximately 400 matched cancer cell lines.

RESULTS

We find that CRISPR screens accurately identify more gene dependencies per cell line, but the majority of each cell line's dependencies are part of a set of 1867 genes that are shared dependencies across the entire collection (pan-lethals). While RNAi knockdown of about 30% of these genes is also pan-lethal, approximately 50% have selective dependency patterns across cell lines, suggesting they could still be cancer vulnerabilities. The accuracy of the unique RNAi selectivity is supported by associations to multi-omics profiles, drug sensitivity, and other expected co-dependencies.

CONCLUSIONS

Incorporating RNAi data for genes that are pan-lethal knockouts facilitates the discovery of a wider range of gene targets than could be detected using the CRISPR dataset alone. This can aid in the interpretation of contrasting results obtained from CRISPR and RNAi screens and reinforce the importance of partial gene suppression methods in building a cancer dependency map.

The landscape of tolerated genetic variation in humans and primates

Personalized genome sequencing has revealed millions of genetic differences between individuals, but our understanding of their clinical relevance remains largely incomplete. To systematically decipher the effects of human genetic variants, we obtained whole-genome sequencing data for 809 individuals from 233 primate species and identified 4.3 million common protein-altering variants with orthologs in humans. We show that these variants can be inferred to have nondeleterious effects in humans based on their presence at high allele frequencies in other primate populations. We use this resource to classify 6% of all possible human protein-altering variants as likely benign and impute the pathogenicity of the remaining 94% of variants with deep learning, achieving state-of-the-art accuracy for diagnosing pathogenic variants in patients with genetic diseases.

Scaling the Discrete-time Wright Fisher model to biobank-scale datasets

The Discrete-Time Wright Fisher (DTWF) model and its large population diffusion limit are central to population genetics. These models describe the forward-in-time evolution of the frequency of an allele in a population and can include the fundamental forces of genetic drift, mutation, and selection. Computing like-lihoods under the diffusion process is feasible, but the diffusion approximation breaks down for large sample sizes or in the presence of strong selection. Unfortunately, existing methods for computing likelihoods under the DTWF model do not scale to current exome sequencing sample sizes in the hundreds of thousands. Here we present an algorithm that approximates the DTWF model with provably bounded error and runs in time linear in the size of the population. Our approach relies on two key observations about Binomial distributions. The first is that Binomial distributions are approximately sparse. The second is that Binomial distributions with similar success probabilities are extremely close as distributions, allowing us to approximate the DTWF Markov transition matrix as a very low rank matrix. Together, these observations enable matrix-vector multiplication in linear (as opposed to the usual quadratic) time. We prove similar properties for Hypergeometric distributions, enabling fast computation of likelihoods for subsamples of the population. We show theoretically and in practice that this approximation is highly accurate and can scale to population sizes in the billions, paving the way for rigorous biobank-scale population genetic inference. Finally, we use our results to estimate how increasing sample sizes will improve the estimation of selection coefficients acting on loss-of-function variants. We find that increasing sample sizes beyond existing large exome sequencing cohorts will provide essentially no additional information except for genes with the most extreme fitness effects.

The landscape of tolerated genetic variation in humans and primates

Personalized genome sequencing has revealed millions of genetic differences between individuals, but our understanding of their clinical relevance remains largely incomplete. To systematically decipher the effects of human genetic variants, we obtained whole genome sequencing data for 809 individuals from 233 primate species, and identified 4.3 million common protein-altering variants with orthologs in human. We show that these variants can be inferred to have non-deleterious effects in human based on their presence at high allele frequencies in other primate populations. We use this resource to classify 6% of all possible human protein-altering variants as likely benign and impute the pathogenicity of the remaining 94% of variants with deep learning, achieving state-of-the-art accuracy for diagnosing pathogenic variants in patients with genetic diseases.

One Sentence Summary

Deep learning classifier trained on 4.3 million common primate missense variants predicts variant pathogenicity in humans.

Inferring the mode and strength of ongoing selection

Genome sequence data are no longer scarce. The UK Biobank alone comprises 200,000 individual genomes, with more on the way, leading the field of human genetics toward sequencing entire populations. Within the next decades, other model organisms will follow suit, especially domesticated species such as crops and livestock. Having sequences from most individuals in a population will present new challenges for using these data to improve health and agriculture in the pursuit of a sustainable future. Existing population genetic methods are designed to model hundreds of randomly sampled sequences but are not optimized for extracting the information contained in the larger and richer data sets that are beginning to emerge, with thousands of closely related individuals. Here we develop a new method called trio-based inference of dominance and selection (TIDES) that uses data from tens of thousands of family trios to make inferences about natural selection acting in a single generation. TIDES further improves on the state of the art by making no assumptions regarding demography, linkage, or dominance. We discuss how our method paves the way for studying natural selection from new angles.

Genomic Underpinnings of Population Persistence in Isle Royale Moose

Island ecosystems provide natural laboratories to assess the impacts of isolation on population persistence. However, most studies of persistence have focused on a single species, without comparisons to other organisms they interact with in the ecosystem. The case study of moose and gray wolves on Isle Royale allows for a direct contrast of genetic variation in isolated populations that have experienced dramatically differing population trajectories over the past decade. Whereas the Isle Royale wolf population recently declined nearly to extinction due to severe inbreeding depression, the moose population has thrived and continues to persist, despite having low genetic diversity and being isolated for ∼120 years. Here, we examine the patterns of genomic variation underlying the continued persistence of the Isle Royale moose population. We document high levels of inbreeding in the population, roughly as high as the wolf population at the time of its decline. However, inbreeding in the moose population manifests in the form of intermediate-length runs of homozygosity suggestive of historical inbreeding and purging, contrasting with the long runs of homozygosity observed in the smaller wolf population. Using simulations, we confirm that substantial purging has likely occurred in the moose population. However, we also document notable increases in genetic load, which could eventually threaten population viability over the long term. Overall, our results demonstrate a complex relationship between inbreeding, genetic diversity, and population viability that highlights the use of genomic datasets and computational simulation tools for understanding the factors enabling persistence in isolated populations.

Relating pathogenic loss-of-function mutations in humans to their evolutionary fitness costs

Causal loss-of-function (LOF) variants for Mendelian and severe complex diseases are enriched in 'mutation intolerant' genes. We show how such observations can be interpreted in light of a model of mutation-selection balance and use the model to relate the pathogenic consequences of LOF mutations at present to their evolutionary fitness effects. To this end, we first infer posterior distributions for the fitness costs of LOF mutations in 17,318 autosomal and 679 X-linked genes from exome sequences in 56,855 individuals. Estimated fitness costs for the loss of a gene copy are typically above 1%; they tend to be largest for X-linked genes, whether or not they have a Y homolog, followed by autosomal genes and genes in the pseudoautosomal region. We compare inferred fitness effects for all possible de novo LOF mutations to those of de novo mutations identified in individuals diagnosed with one of six severe, complex diseases or developmental disorders. Probands carry an excess of mutations with estimated fitness effects above 10%; as we show by simulation, when sampled in the population, such highly deleterious mutations are typically only a couple of generations old. Moreover, the proportion of highly deleterious mutations carried by probands reflects the typical age of onset of the disease. The study design also has a discernible influence: a greater proportion of highly deleterious mutations is detected in pedigree than case-control studies, and for autism, in simplex than multiplex families and in female versus male probands. Thus, anchoring observations in human genetics to a population genetic model allows us to learn about the fitness effects of mutations identified by different mapping strategies and for different traits.

Molecular Landscape of Tourette's Disorder

Tourette's disorder (TD) is a highly heritable childhood-onset neurodevelopmental disorder and is caused by a complex interplay of multiple genetic and environmental factors. Yet, the molecular mechanisms underlying the disorder remain largely elusive. In this study, we used the available omics data to compile a list of TD candidate genes, and we subsequently conducted tissue/cell type specificity and functional enrichment analyses of this list. Using genomic data, we also investigated genetic sharing between TD and blood and cerebrospinal fluid (CSF) metabolite levels. Lastly, we built a molecular landscape of TD through integrating the results from these analyses with an extensive literature search to identify the interactions between the TD candidate genes/proteins and metabolites. We found evidence for an enriched expression of the TD candidate genes in four brain regions and the pituitary. The functional enrichment analyses implicated two pathways ('cAMP-mediated signaling' and 'Endocannabinoid Neuronal Synapse Pathway') and multiple biological functions related to brain development and synaptic transmission in TD etiology. Furthermore, we found genetic sharing between TD and the blood and CSF levels of 39 metabolites. The landscape of TD not only provides insights into the (altered) molecular processes that underlie the disease but, through the identification of potential drug targets (such as FLT3, NAALAD2, CX3CL1-CX3CR1, OPRM1, and HRH2), it also yields clues for developing novel TD treatments.

Informing variant assessment using structured evidence from prior classifications (PS1, PM5, and PVS1 sequence variant interpretation criteria)

PURPOSE

This study aimed to explore whether evidence of pathogenicity from prior variant classifications in ClinVar could be used to inform variant interpretation using the American College of Medical Genetics and Genomics/Association for Molecular Pathology clinical guidelines.

METHODS

We identified distinct single-nucleotide variants (SNVs) that are either similar in location or in functional consequence to pathogenic variants in ClinVar and analyzed evidence in support of pathogenicity using 3 interpretation criteria.

RESULTS

Thousands of variants, including many in clinically actionable disease genes (American College of Medical Genetics and Genomics secondary findings v3.0), have evidence of pathogenicity from existing variant classifications, accounting for 2.5% of nonsynonymous SNVs within ClinVar. Notably, there are many variants with uncertain or conflicting classifications that cause the same amino acid substitution as other pathogenic variants (PS1, N = 323), variants that are predicted to cause different amino acid substitutions in the same codon as pathogenic variants (PM5, N = 7692), and loss-of-function variants that are present in genes in which many loss-of-function variants are classified as pathogenic (PVS1, N = 3635). Most of these variants have similar computational predictions of pathogenicity and splicing effect as their associated pathogenic variants.

CONCLUSION

Broadly, for >1.4 million SNVs exome wide, information from previously classified variants could be used to provide evidence of pathogenicity. We have developed a pipeline to identify variants meeting these criteria that may inform interpretation efforts.

The molecular evolution of spermatogenesis across mammals

The testis produces gametes through spermatogenesis and evolves rapidly at both the morphological and molecular level in mammals^1-6, probably owing to the evolutionary pressure on males to be reproductively successful⁷. However, the molecular evolution of individual spermatogenic cell types across mammals remains largely uncharacterized. Here we report evolutionary analyses of single-nucleus transcriptome data for testes from 11 species that cover the three main mammalian lineages (eutherians, marsupials and monotremes) and birds (the evolutionary outgroup), and include seven primates. We find that the rapid evolution of the testis was driven by accelerated fixation rates of gene expression changes, amino acid substitutions and new genes in late spermatogenic stages, probably facilitated by reduced pleiotropic constraints, haploid selection and transcriptionally permissive chromatin. We identify temporal expression changes of individual genes across species and conserved expression programs controlling ancestral spermatogenic processes. Genes predominantly expressed in spermatogonia (germ cells fuelling spermatogenesis) and Sertoli (somatic support) cells accumulated on X chromosomes during evolution, presumably owing to male-beneficial selective forces. Further work identified transcriptomal differences between X- and Y-bearing spermatids and uncovered that meiotic sex-chromosome inactivation (MSCI) also occurs in monotremes and hence is common to mammalian sex-chromosome systems. Thus, the mechanism of meiotic silencing of unsynapsed chromatin, which underlies MSCI, is an ancestral mammalian feature. Our study illuminates the molecular evolution of spermatogenesis and associated selective forces, and provides a resource for investigating the biology of the testis across mammals.

Biobanking as a Tool for Genomic Research: From Allele Frequencies to Cross-Ancestry Association Studies

In recent years, great advances have been made in the field of collection, storage, and analysis of biological samples. Large collections of samples, biobanks, have been established in many countries. Biobanks typically collect large amounts of biological samples and associated clinical information; the largest collections include over a million samples. In this review, we summarize the main directions in which biobanks aid medical genetics and genomic research, from providing reference allele frequency information to allowing large-scale cross-ancestry meta-analyses. The largest biobanks greatly vary in the size of the collection, and the amount of available phenotype and genotype data. Nevertheless, all of them are extensively used in genomics, providing a rich resource for genome-wide association analysis, genetic epidemiology, and statistical research into the structure, function, and evolution of the human genome. Recently, multiple research efforts were based on trans-biobank data integration, which increases sample size and allows for the identification of robust genetic associations. We provide prominent examples of such data integration and discuss important caveats which have to be taken into account in trans-biobank research.

Combining genetic constraint with predictions of alternative splicing to prioritize deleterious splicing in rare disease studies

Background

Despite numerous molecular and computational advances, roughly half of patients with a rare disease remain undiagnosed after exome or genome sequencing. A particularly challenging barrier to diagnosis is identifying variants that cause deleterious alternative splicing at intronic or exonic loci outside of canonical donor or acceptor splice sites.

Results

Several existing tools predict the likelihood that a genetic variant causes alternative splicing. We sought to extend such methods by developing a new metric that aids in discerning whether a genetic variant leads to deleterious alternative splicing. Our metric combines genetic variation in the Genome Aggregate Database with alternative splicing predictions from SpliceAI to compare observed and expected levels of splice-altering genetic variation. We infer genic regions with significantly less splice-altering variation than expected to be constrained. The resulting model of regional splicing constraint captures differential splicing constraint across gene and exon categories, and the most constrained genic regions are enriched for pathogenic splice-altering variants. Building from this model, we developed ConSpliceML. This ensemble machine learning approach combines regional splicing constraint with multiple per-nucleotide alternative splicing scores to guide the prediction of deleterious splicing variants in protein-coding genes. ConSpliceML more accurately distinguishes deleterious and benign splicing variants than state-of-the-art splicing prediction methods, especially in “cryptic” splicing regions beyond canonical donor or acceptor splice sites.

Conclusion

Integrating a model of genetic constraint with annotations from existing alternative splicing tools allows ConSpliceML to prioritize potentially deleterious splice-altering variants in studies of rare human diseases.

Predicting functional effect of missense variants using graph attention neural networks

Accurate prediction of damaging missense variants is critically important for interpreting a genome sequence. Although many methods have been developed, their performance has been limited. Recent advances in machine learning and the availability of large-scale population genomic sequencing data provide new opportunities to considerably improve computational predictions. Here we describe the graphical missense variant pathogenicity predictor (gMVP), a new method based on graph attention neural networks. Its main component is a graph with nodes that capture predictive features of amino acids and edges weighted by co-evolution strength, enabling effective pooling of information from the local protein context and functionally correlated distal positions. Evaluation of deep mutational scan data shows that gMVP outperforms other published methods in identifying damaging variants in TP53, PTEN, BRCA1 and MSH2. Furthermore, it achieves the best separation of de novo missense variants in neuro developmental disorder cases from those in controls. Finally, the model supports transfer learning to optimize gain- and loss-of-function predictions in sodium and calcium channels. In summary, we demonstrate that gMVP can improve interpretation of missense variants in clinical testing and genetic studies.

Evolution and dysfunction of human cognitive and social traits: A transcriptional regulation perspective

Evolutionary changes in brain and craniofacial development have endowed humans with unique cognitive and social skills, but also predisposed us to debilitating disorders in which these traits are disrupted. What are the developmental genetic underpinnings that connect the adaptive evolution of our cognition and sociality with the persistence of mental disorders with severe negative fitness effects? We argue that loss of function of genes involved in transcriptional regulation represents a crucial link between the evolution and dysfunction of human cognitive and social traits. The argument is based on the haploinsufficiency of many transcriptional regulator genes, which makes them particularly sensitive to loss-of-function mutations. We discuss how human brain and craniofacial traits evolved through partial loss of function (i.e. reduced expression) of these genes, a perspective compatible with the idea of human self-domestication. Moreover, we explain why selection against loss-of-function variants supports the view that mutation-selection-drift, rather than balancing selection, underlies the persistence of psychiatric disorders. Finally, we discuss testable predictions.

Integrating de novo and inherited variants in 42,607 autism cases identifies mutations in new moderate-risk genes

To capture the full spectrum of genetic risk for autism, we performed a two-stage analysis of rare de novo and inherited coding variants in 42,607 autism cases, including 35,130 new cases recruited online by SPARK. We identified 60 genes with exome-wide significance (P < 2.5 × 10^-6), including five new risk genes (NAV3, ITSN1, MARK2, SCAF1 and HNRNPUL2). The association of NAV3 with autism risk is primarily driven by rare inherited loss-of-function (LoF) variants, with an estimated relative risk of 4, consistent with moderate effect. Autistic individuals with LoF variants in the four moderate-risk genes (NAV3, ITSN1, SCAF1 and HNRNPUL2; n = 95) have less cognitive impairment than 129 autistic individuals with LoF variants in highly penetrant genes (CHD8, SCN2A, ADNP, FOXP1 and SHANK3) (59% vs 88%, P = 1.9 × 10^-6). Power calculations suggest that much larger numbers of autism cases are needed to identify additional moderate-risk genes.

Most cancers carry a substantial deleterious load due to Hill-Robertson interference

Cancer genomes exhibit surprisingly weak signatures of negative selection^1,2. This may be because selective pressures are relaxed or because genome-wide linkage prevents deleterious mutations from being removed (Hill-Robertson interference)³. By stratifying tumors by their genome-wide mutational burden, we observe negative selection (dN/dS ~ 0.56) in low mutational burden tumors, while remaining cancers exhibit dN/dS ratios ~1. This suggests that most tumors do not remove deleterious passengers. To buffer against deleterious passengers, tumors upregulate heat shock pathways as their mutational burden increases. Finally, evolutionary modeling finds that Hill-Robertson interference alone can reproduce patterns of attenuated selection and estimates the total fitness cost of passengers to be 46% per cell on average. Collectively, our findings suggest that the lack of observed negative selection in most tumors is not due to relaxed selective pressures, but rather the inability of selection to remove deleterious mutations in the presence of genome-wide linkage.

A cross-disorder dosage sensitivity map of the human genome

Rare copy-number variants (rCNVs) include deletions and duplications that occur infrequently in the global human population and can confer substantial risk for disease. In this study, we aimed to quantify the properties of haploinsufficiency (i.e., deletion intolerance) and triplosensitivity (i.e., duplication intolerance) throughout the human genome. We harmonized and meta-analyzed rCNVs from nearly one million individuals to construct a genome-wide catalog of dosage sensitivity across 54 disorders, which defined 163 dosage sensitive segments associated with at least one disorder. These segments were typically gene dense and often harbored dominant dosage sensitive driver genes, which we were able to prioritize using statistical fine-mapping. Finally, we designed an ensemble machine-learning model to predict probabilities of dosage sensitivity (pHaplo & pTriplo) for all autosomal genes, which identified 2,987 haploinsufficient and 1,559 triplosensitive genes, including 648 that were uniquely triplosensitive. This dosage sensitivity resource will provide broad utility for human disease research and clinical genetics.

The impact of genetic modifiers on variation in germline mutation rates within and among human populations

Mutation rates and spectra differ among human populations. Here, we examine whether this variation could be explained by evolution at mutation modifiers. To this end, we consider genetic modifier sites at which mutations, "mutator alleles," increase genome-wide mutation rates and model their evolution under purifying selection due to the additional deleterious mutations that they cause, genetic drift, and demographic processes. We solve the model analytically for a constant population size and characterize how evolution at modifier sites impacts variation in mutation rates within and among populations. We then use simulations to study the effects of modifier sites under a plausible demographic model for Africans and Europeans. When comparing populations that evolve independently, weakly selected modifier sites (2Nes≈1), which evolve slowly, contribute the most to variation in mutation rates. In contrast, when populations recently split from a common ancestral population, strongly selected modifier sites (2Nes≫1), which evolve rapidly, contribute the most to variation between them. Moreover, a modest number of modifier sites (e.g. 10 per mutation type in the standard classification into 96 types) subject to moderate to strong selection (2Nes>1) could account for the variation in mutation rates observed among human populations. If such modifier sites indeed underlie differences among populations, they should also cause variation in mutation rates within populations and their effects should be detectable in pedigree studies.

The arginine methyltransferase Carm1 is necessary for heart development

To discover genes implicated in human congenital disorders, we performed ENU mutagenesis in the mouse and screened for mutations affecting embryonic development. In this work, we report defects of heart development in mice homozygous for a mutation of coactivator-associated arginine methyltransferase 1 (Carm1). While Carm1 has been extensively studied, it has never been previously associated with a role in heart development. Phenotype analysis combining histology and microcomputed tomography imaging shows a range of cardiac defects. Most notably, many affected midgestation embryos appear to have cardiac rupture and hemorrhaging in the thorax. Mice that survive to late gestation show a variety of cardiac defects, including ventricular septal defects, double outlet right ventricle, and persistent truncus arteriosus. Transcriptome analyses of the mutant embryos by mRNA-seq reveal the perturbation of several genes involved in cardiac morphogenesis and muscle development and function. In addition, we observe the mislocalization of cardiac neural crest cells at E12.5 in the outflow tract. The cardiac phenotype of Carm1 mutant embryos is similar to that of Pax3 null mutants, and PAX3 is a putative target of CARM1. However, our analysis does not support the hypothesis that developmental defects in Carm1 mutant embryos are primarily due to a functional defect of PAX3.

Extreme purifying selection against point mutations in the human genome

Large-scale genome sequencing has enabled the measurement of strong purifying selection in protein-coding genes. Here we describe a new method, called ExtRaINSIGHT, for measuring such selection in noncoding as well as coding regions of the human genome. ExtRaINSIGHT estimates the prevalence of "ultraselection" by the fractional depletion of rare single-nucleotide variants, after controlling for variation in mutation rates. Applying ExtRaINSIGHT to 71,702 whole genome sequences from gnomAD v3, we find abundant ultraselection in evolutionarily ancient miRNAs and neuronal protein-coding genes, as well as at splice sites. By contrast, we find much less ultraselection in other noncoding RNAs and transcription factor binding sites, and only modest levels in ultraconserved elements. We estimate that ~0.4-0.7% of the human genome is ultraselected, implying ~ 0.26-0.51 strongly deleterious mutations per generation. Overall, our study sheds new light on the genome-wide distribution of fitness effects by combining deep sequencing data and classical theory from population genetics.

Sporadic hypothalamic hamartoma is a ciliopathy with somatic and bi-allelic contributions

Hypothalamic hamartoma with gelastic seizures is a well-established cause of drug-resistant epilepsy in early life. The development of novel surgical techniques has permitted the genomic interrogation of hypothalamic hamartoma tissue. This has revealed causative mosaic variants within GLI3, OFD1 and other key regulators of the sonic-hedgehog pathway in a minority of cases. Sonic-hedgehog signalling proteins localize to the cellular organelle primary cilia. We therefore explored the hypothesis that cilia gene variants may underlie hitherto unsolved cases of sporadic hypothalamic hamartoma. We performed high-depth exome sequencing and chromosomal microarray on surgically resected hypothalamic hamartoma tissue and paired leukocyte-derived DNA from 27 patients. We searched for both germline and somatic variants under both dominant and bi-allelic genetic models. In hamartoma-derived DNA of seven patients we identified bi-allelic (one germline, one somatic) variants within one of four cilia genes-DYNC2I1, DYNC2H1, IFT140 or SMO. In eight patients, we identified single somatic variants in the previously established hypothalamic hamartoma disease genes GLI3 or OFD1. Overall, we established a plausible molecular cause for 15/27 (56%) patients. Here, we expand the genetic architecture beyond single variants within dominant disease genes that cause sporadic hypothalamic hamartoma to bi-allelic (one germline/one somatic) variants, implicate three novel cilia genes and reconceptualize the disorder as a ciliopathy.

Cancer Relevance of Human Genes

BACKGROUND

We hypothesize that genes that directly or indirectly interact with core cancer genes (CCGs) in a comprehensive gene-gene interaction network may have functional importance in cancer.

METHODS

We categorized 12 767 human genes into CCGs (n = 468), 1 (n = 5467), 2 (n = 5573), 3 (n = 915), and more than 3 steps (n = 416) removed from the nearest CCG in the Search Tool for the Retrieval of Interacting Genes/Proteins network. We estimated cancer-relevant functional importance in these neighborhood categories using 1) gene dependency score, which reflects the effect of a gene on cell viability after knockdown; 2) somatic mutation frequency in The Cancer Genome Atlas; 3) effect size that estimates to what extent a mutation in a gene enhances cell survival; and 4) negative selection pressure of germline protein-truncating variants in healthy populations.

RESULTS

Cancer biology-related functional importance of genes decreases as their distance from the CCGs increases. Genes closer to cancer genes show greater connectedness in the network, have greater importance in maintaining cancer cell viability, are under greater negative germline selection pressure, and have higher somatic mutation frequency in cancer. Based on these 4 metrics, we provide cancer relevance annotation to known human genes.

CONCLUSIONS

A large number of human genes are connected to CCGs and could influence cancer biology to various extent when dysregulated; any given mutation may be functionally important in one but not in another individual depending on genomic context.

Whole genome sequencing delineates regulatory, copy number, and cryptic splice variants in early onset cardiomyopathy

Cardiomyopathy (CMP) is a heritable disorder. Over 50% of cases are gene-elusive on clinical gene panel testing. The contribution of variants in non-coding DNA elements that result in cryptic splicing and regulate gene expression has not been explored. We analyzed whole-genome sequencing (WGS) data in a discovery cohort of 209 pediatric CMP patients and 1953 independent replication genomes and exomes. We searched for protein-coding variants, and non-coding variants predicted to affect the function or expression of genes. Thirty-nine percent of cases harbored pathogenic coding variants in known CMP genes, and 5% harbored high-risk loss-of-function (LoF) variants in additional candidate CMP genes. Fifteen percent harbored high-risk regulatory variants in promoters and enhancers of CMP genes (odds ratio 2.25, p = 6.70 × 10-7 versus controls). Genes involved in α-dystroglycan glycosylation (FKTN, DTNA) and desmosomal signaling (DSC2, DSG2) were most highly enriched for regulatory variants (odds ratio 6.7-58.1). Functional effects were confirmed in patient myocardium and reporter assays in human cardiomyocytes, and in zebrafish CRISPR knockouts. We provide strong evidence for the genomic contribution of functionally active variants in new genes and in regulatory elements of known CMP genes to early onset CMP.

Towards SINEUP-based therapeutics: Design of an in vitro synthesized SINEUP RNA

SINEUPs are a novel class of natural and synthetic non-coding antisense RNA molecules able to increase the translation of a target mRNA. They present a modular organization comprising an unstructured antisense target-specific domain, which sets the specificity of each individual SINEUP, and a structured effector domain, which is responsible for the translation enhancement. In order to design a fully functional in vitro transcribed SINEUP for therapeutics applications, SINEUP RNAs were synthesized in vitro with a variety of chemical modifications and screened for their activity on endogenous target mRNA upon transfection. Three combinations of modified ribonucleotides-2'O methyl-ATP (Am), N6 methyl-ATP (m6A), and pseudo-UTP (ψ)-conferred SINEUP activity to naked RNA. The best combination tested in this study was fully modified with m6A and ψ. Aside from functionality, this combination conferred improved stability upon transfection and higher thermal stability. Common structural determinants of activity were identified by circular dichroisms, defining a core functional structure that is achieved with different combinations of modifications.

Leveraging gene co-regulation to identify gene sets enriched for disease heritability

Identifying gene sets that are associated to disease can provide valuable biological knowledge, but a fundamental challenge of gene set analyses of GWAS data is linking disease-associated SNPs to genes. Transcriptome-wide association studies (TWASs) detect associations between the genetically predicted expression of a gene and disease risk, thus implicating candidate disease genes. However, causal disease genes at TWAS-associated loci generally remain unknown due to gene co-regulation, which leads to correlations across genes in predicted expression. We developed a method, gene co-regulation score (GCSC) regression, to identify gene sets that are enriched for disease heritability explained by predicted expression. GCSC regresses TWAS chi-square statistics on gene co-regulation scores reflecting correlations in predicted gene expression; a gene set is enriched for heritability if genes with high co-regulation to the set have higher TWAS chi-square statistics than genes with low co-regulation to the set, beyond what is expected based on co-regulation to all genes. We verified via simulations that GCSC is well calibrated and well powered. We applied GCSC to gene expression data from GTEx (48 tissues) and GWAS summary statistics for 43 independent diseases and complex traits analyzing a broad set of biological pathways and specifically expressed gene sets. We identified many enriched sets, recapitulating known biology. For Alzheimer disease, we detected evidence of an immune basis, and specifically a role for antigen presentation, in analyses of both biological pathways and specifically expressed gene sets. Our results highlight the advantages of leveraging gene co-regulation within the TWAS framework to identify enriched gene sets.

Reduced reproductive success is associated with selective constraint on human genes

Genome-wide sequencing of human populations has revealed substantial variation among genes in the intensity of purifying selection acting on damaging genetic variants¹. Although genes under the strongest selective constraint are highly enriched for associations with Mendelian disorders, most of these genes are not associated with disease and therefore the nature of the selection acting on them is not known². Here we show that genetic variants that damage these genes are associated with markedly reduced reproductive success, primarily owing to increased childlessness, with a stronger effect in males than in females. We present evidence that increased childlessness is probably mediated by genetically associated cognitive and behavioural traits, which may mean that male carriers are less likely to find reproductive partners. This reduction in reproductive success may account for 20% of purifying selection against heterozygous variants that ablate protein-coding genes. Although this genetic association may only account for a very minor fraction of the overall likelihood of being childless (less than 1%), especially when compared to more influential sociodemographic factors, it may influence how genes evolve over time.

Pairs of mutually compensatory frameshifting mutations contribute to protein evolution

Insertions and deletions of lengths not divisible by 3 in protein-coding sequences cause frameshifts that usually induce premature stop codons and may carry a high fitness cost. However, this cost can be partially offset by a second compensatory indel restoring the reading frame. The role of such pairs of compensatory frameshifting mutations (pCFMs) in evolution has not been studied systematically. Here, we use whole-genome alignments of protein-coding genes of 100 vertebrate species, and of 122 insect species, studying the prevalence of pCFMs in their divergence. We detect a total of 624 candidate pCFM genes; six of them pass stringent quality filtering, including three human genes: RAB36, ARHGAP6, and NCR3LG1. In some instances, amino acid substitutions closely predating or following pCFMs restored the biochemical similarity of the frameshifted segment to the ancestral amino acid sequence, possibly reducing or negating the fitness cost of the pCFM. Typically, however, the biochemical similarity of the frameshifted sequence to the ancestral one was not higher than the similarity of a random sequence of a protein-coding gene to its frameshifted version, indicating that pCFMs can uncover radically novel regions of protein space. In total, pCFMs represent an appreciable and previously overlooked source of novel variation in amino acid sequences.

No preferential mode of inheritance for highly constrained genes

Genetic constraint metrics such as the gnomAD probability of being loss-of-function (LoF) intolerant (pLI) are used to prioritize candidate genes but the mode of inheritance of highly constrained genes has never specifically been studied. We compared 605 genes with a pLI of 1 (pLI1 group) with a random sample of 635 genes from gnomAD (the random group) in terms of genetic constraint metrics, associations with Mendelian disease, modes of inheritance, and two intragenic constraint scores: the percentage of constraint coding regions (CCR) in the 99^th percentile and the gene variation intolerance rank (GeVIR). The proportion of genes associated with a Mendelian disease was 35.9% (217/605) in the pLI1 group and 19.5% (124/635) in the random group (p < 0.0001). The modes of inheritance in the random group were autosomal dominant for 35 genes (28.2%), autosomal recessive for 69 (55.6%), mixed for 14 (11.3%) and X-linked for 6 genes (4.8%). The corresponding distribution in the pLI1 group was 150 (69.1%), 26 (12.0%), 14 (6.5%) and 27 (12.4%) (p < 0.0001). The percentage of CCRs in the 99th percentile was 0.3 in the random group versus 1.12 in the pLI1 group (p < 0.0001). The GeVIR score was 50.9 for the random group versus 15.1 for the pLI1 group (p < 0.0001). High genetic constraint does not seem to be associated with a particular mode of inheritance but does seem to be associated with the intragenic constraint scores considered here. Some highly constrained genes are associated with two different modes of inheritance.