NHGRI - National Human Genome Research Institute Grants

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY The decreasing cost of genomic sequencing will yield millions of samples in the coming years from both research and healthcare. Sharing this data is necessary to understand human diseases and eventually help patients, but doing so requires the community to agree on common methods for collecting, storing, transferring, accessing, and analyzing data. This proposal will support the Global Alliance for Genomics and Health (GA4GH) to aid genomic research and human health by developing standards and policies for effective and responsible data sharing between institutions and countries around the world. To advance responsible sharing of global genomic and health-related data, genomics researchers, clinicians, bioinformaticians, software engineers, and industry experts will work together as a single GA4GH community to deliver genomic data sharing standards and frameworks (e.g., ontologies, guidelines, technical schemas). Building on our five years of experience convening stakeholders and developing work products, we will engage the genomics and health community in the very earliest stages of development to ensure our work is useful and ready for adoption. We will leverage the combined effort of several hundred active contributors to advance development activities beyond the capacity of our small staff team. These contributors will work within eight GA4GH Work Streams, each focused on developing critical standards and frameworks, including cloud-based data federation, scalable schemas and interfaces, data models, and file formats. We will engage deeply within the broader healthcare, research, and commercial sectors, including the launch of the Genomics in Health Implementation Forum to drive uptake in the clinical domain. A federated ecosystem for searching, discovering, exchanging, and analyzing genomic and clinical data will enable a global learning health system that advances both research and clinical care beyond their individual capacities and depends on standards and interoperable frameworks embraced by the entire community. We envision a future in which the full suite of GA4GH standards enables all clinicians, geneticists, and researchers to search across the world’s collective genomic data to reveal unanticipated gene-disease associations, make otherwise impossible drug-response predictions, and generally participate in genomics at a competitive pace—regardless of their means or location. The promise of genomic medicine lies at a crossroads that depends on harmonization across the community and will significantly enhance the human experience if we succeed. We believe that GA4GH is necessary to that success.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY Despite significant advancements in genomic medicine, its integration into Internal Medicine (IM) remains limited. This program will seek to address barriers at the residency training level. Mount Sinai has been at the forefront of addressing this need for genomic medicine training during Internal Medicine training. Mount Sinai has stood up initiatives such as a specialized Genomic Medicine track within the IM residency and the development of the Genomic Education in Medicine (GEM) platform, which provides web-based resources to support genomic education for IM residents. This one-day conference will bring together leaders from IM residency programs across the United States, as well as genomic medicine experts and clinicians to assess current efforts and share innovations for integrating genomics into training. The event will focus on identifying barriers to adoption, such as limited expertise and competing priorities, and developing strategies to overcome these challenges across diverse residency environments. Through a collaborative one-day process, attendees will create a “consensus agenda for education in and implementation of genomic medicine” within IM residencies, which will then be refined and finalized through follow-up virtual meetings. The goal is to publish a proposed education agenda and implementation strategy to support those IM training programs that seek to bring genomic medicine implementation into their curriculums.

NHGRI - National Human Genome Research Institute

Abstract Large-scale genome wide association studies (GWAS) have identified a large number of genetic variants associated with complex diseases. The aggregation of all the variants that are known to contribute to the disease in the form of polygenic risk scores (PRS) improves the prediction of a range of complex diseases. Most PRS have been developed within European ancestry study samples and have shown to perform poorly in other ancestry groups, further exaggerating differences in treatment quality across ancestries. As genetic approaches for precision medicine become more popular, there is a critical need to responsively and proactively expand access to accurate PRS. Specifically, diabetes, and its associated complications are one of the biggest health problems of the 21st century. In fact, type 1 and type 2 diabetes (T1D and T2D), gestational diabetes (GDM) and related complications are excellent disease models to study the utility of PRS for predicting heterogenous and complex health outcomes in a setting where differences in access to healthcare exist. Not only are PRS useful to predict T1D and T2D, but they can distinguish between T1D and T2D, and between T2D subtypes. The wealth of existing GWAS data from diabetes subtypes, complications, and quantitative traits recently generated provides a unique opportunity for constructing highly transferable PRS across populations. To address the differences in accuracy in PRS across ancestries, we have assembled a multi-disciplinary team to aggregate and analyze the largest existing genetic data from more than 1.8 M individuals (49% non-European) with T1D, T2D, GDM and glycemia-related complications and quantitative traits to improve the PRS prediction of diabetes and progression across lifespan in multiple ancestries with these Aims: (1) Collection, harmonization and integration of large-scale, multi-ancestry cohorts with diabetes traits across the life-span and genomics for development, training and testing PRS for multiple ancestries; (2) Development of methods to improve PRS prediction in non-European populations by using Bayesian approaches that allow integration of linkage disequilibrium and summary statistics from several ancestries. (3) Development, testing, and comparing performance of PRS for each trait, development of risk prediction tools that integrate clinical and genetic risk factors, and assessment of scenarios where PRS improve the prediction. Accomplishing the aims of this proposal will demonstrate how genomic data can inform more efficient and targeted preventive strategies within healthcare systems and across multiple populations. Findings are expected to advance precision care of patients with diabetes and related conditions in people of multiple ancestral backgrounds and serve as a paradigm for many other complex diseases.

NHGRI - National Human Genome Research Institute

The supplement instructions PA-26-001 state: “At a minimum, the Research Strategy or Program Plan section, as applicable, must be completed and must include a summary or abstract of the funded parent award or project. Other sections should also be included if they are being changed by the proposed supplement activities.” The original Project Summary/Abstract document has not changed as a result of the proposed supplement activities.

NHGRI - National Human Genome Research Institute

Abstract The Polygenic Risk Methods Development (PRIMED) Consortium was established to harmonize large-scale genomic and phenotypic data and advance methodological innovation to improve polygenic risk scores (PRS). Within the scope of the PRIMED Consortium, the parent award: Functional and Fine-Mapping Approach to Improve Responsible Risk-modeling of Polygenic Risk Scores (FFAIRR-PRS, U01HG011719) aimed to harmonize and aggregate individual-level data across populations, to share analytic workflows and new statistical genetics frameworks, and to validate these approaches for enhanced polygenic score development, generalizability, and clinical translation. To that end, our team has developed new methods with requisite validation advancing new generalizable frameworks for polygenic risk scoring that have already translated to clinical adoption. Based on these new frameworks and consortium-wide learnings, the goal of this administrative supplement is to complete several ongoing high-impact consortium-wide and site-specific activities that will not yet be finalized within the current project period toward maximal impact for PRIMED. Our research team combines strengths in cardiovascular medicine, statistical genetics, and high throughput genetics and genomics. In Aim 1, we will combine existing PRS from the PGS Catalog using the PRSmix method to enhance cross-trait prediction accuracy and stability and will examine the scale of risk estimation and contextualization effects to optimize fairness in trait-specific PRS modeling across contexts. In Aim 2, we will finalize and disseminate new methodological frameworks, including integrative approaches that involve individual-specific trajectories and rare variant modeling. In Aim 3, we will extend benchmarking and integrative risk modeling efforts across key traits to improve clinical translation. Completion of our aims will further advance PRS accuracy and generalizability and enable methodological advances. We will integrate our learnings to both describe the current state of the field but also provide granular guidance regarding clinical translation and communication with patients.

NHGRI - National Human Genome Research Institute

Polygenic Risk Scores (PRS) developed from analysis of large-scale genomic data in epidemiological studies hold promise as a precision medicine tool that will help identify individuals at higher disease risk. However, a critical challenge now facing PRS research and clinical translation is that most PRS have been developed using data from individuals in limited populations, resulting in poorer predictive performance across populations and reflecting the lack of representation of genetic variation in genomic and biomedical research more broadly. NHGRI is establishing a new Polygenic Risk Score Methods Development Consortium to address this challenge by (1) leveraging existing cohorts to improve PRS prediction across populations and for a range of conditions and (2) optimizing the integration of large-scale genomic and phenotype datasets in support of collaborative analysis, reporting, and creation of methods and resources for the broader scientific community. As Coordinating Center, we will support the Consortium's goals by achieving four main aims: (1) leading harmonization of genotype and phenotype data across Consortium Study Sites and Affiliate Members, including variant and sample level quality control, genotype imputation, and use of standard phenotype ontologies. (2) Organizing cross-Consortium analysis and collaborative methods development, featuring standardized evaluation of PRS methods to identify consensus approaches, development of a local-ancestry informed PRS method, and integration of ELSI considerations into analytic best-practices. (3) Facilitating data sharing within and beyond the Consortium through community resources and repositories such as the NHGRI Genomic Data Science Analysis, Visualization, and Informatics Lab-space (AnVIL), NCBI database of Genotypes and Phenotypes (dbGaP), and the Polygenic Score Catalog. (4) Coordinating program logistics and outreach including through a Consortium website, in-person meetings, mentorship and training, and regular engagement with precision medicine partner programs. We propose innovative approaches to achieve these four main aims, including leveraging emerging cloud-based platforms for harmonization, analysis, and sharing of large-scale genomic data; investigating scalable data science approaches to phenotype harmonization; leading a Consortium-wide “bake-off” to establish consensus approaches to PRS generation and evaluation; and incorporating Ethical, Legal, and Social Implications into analytic best practices. Furthermore, our application rests on 13 successful years of experience and expertise serving as Coordinating Centers for five large-scale genetic and biomedical projects. Through scientific and administrative leadership of this Consortium, we will help realize the public health benefit of developing PRS that predict and help prevent or mitigate a range of diseases.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY This administrative supplement request seeks additional support for analysis and data sharing within the GREGoR consortium under parent grant U01HG011755, “Broad Institute Mendelian Genomic Research Center.” Although GREGoR sites have generated extensive long-read (lrGS) and short-read genome sequencing (srGS) data, heterogeneous methodologies—including different technologies, chemistries, read depths, and analysis pipelines—have resulted in fragmented datasets, limiting cross-cohort discovery and standardized evaluation of the diagnostic yield generated in rare disease studies. This proposal leverages cloud-based, standardized pipelines (as employed by the All of Us Research Program and HGSVC) to harmonize lrGS data across >1,500 individuals, generates a uniform long-read callset, and integrates these findings with a joint structural variant (SV) callset from srGS data and complementary multiomics analyses (including DNA methylation data). We will also develop the seqr platform to expand controlled-access sharing of the GREGoR dataset. Finally, we will complete analysis of the samples that have been sequenced to date in GREGoR by applying a variant prioritization algorithm and generating a structural variant callset from the short read genome data. We will also deploy novel methods in transcriptomic analysis, noncoding variant prioritization, and tandem repeat expansion discovery. Results will continue to be returned to participants, and individual gene-disease relationships and variant classifications will be shared through GenCC and ClinVar, respectively. Our approaches will work towards achieving the GREGoR consortium’s goals of enabling robust joint analyses across GREGoR sites, providing a rigorously annotated, controlled-access resource designed to maximize diagnostic yield, benchmarking novel variant discoveries, and facilitating widespread innovation in the application of short-read and long-read sequencing technologies for rare disease research.

NHGRI - National Human Genome Research Institute

Abstract In the United States, 60% of adults have a chronic disease, and 42% have two or more, accounting for over 90% of the nation’s $4.5 trillion in annual healthcare expenditures. Incorporating genomic risk (polygenic risk scores (PRS) and monogenic risk), and clinical and family health history (FHHx) data into a genome-informed risk assessment (GIRA) can support proactive healthcare by identifying those at elevated risk who may benefit from earlier screening, preventive strategies, or timely initiation or intensification of treatment to delay, reduce, or prevent disease onset. The electronic MEdical Records and GEnomics (eMERGE) network brought together expertise and experience of investigators at 10 sites to develop and validate PRS for 10 chronic conditions. GIRA reports, communicating PRS, monogenic, clinical, and FHHx-based risk for 11 conditions (Asthma, Atrial fibrillation, Breast Cancer, Chronic Kidney Disease, Colorectal Cancer, Coronary Heart Disease, Hypercholesterolemia, Obesity, Prostate Cancer, Type 1 and Type 2 Diabetes) were returned to 23,840 participants (19,535 adults and 4305 children), including risk-reduction recommendations for high-risk individuals. The University of Alabama at Birmingham (UAB) has led network-wide outcomes assessment, data extraction and harmonization, and statistical analyses and actively collaborated with the eMERGE network developing PRS for Chronic Kidney Disease and Type 2 Diabetes, enrolled and completed GIRA returns to 2927 participants (2871 adults, 56 children). All GIRAs were returned to the provider, and participants and uploaded to the EHR. We have completed provider and participant surveys and extracted 6-month EHR outcomes. Participants have completed 12-month follow-up. Data extraction, quality control and harmonization of 12- month outcomes is underway with expected completion by April 2026. This one-year extension (Year 7) request will allow UAB eMERGE IV site to QC and harmonize 12-month outcome data, analyze whether returning high risk GIRA leads to new preventive and screening healthcare actions, and disseminate results through manuscripts. UAB will lead the outcomes assessment and assess provider and participant adoption of recommendations using 12-month data. We will collaborate with the emerge investigative team and NHGRI leadership to continue to discover and publish the findings from the eMERGE IV study.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY The electronic MEdical Records and GEnomics (eMERGE IV) Network is a national network organized and funded by the National Human Genome Research Institute (NHGRI) that combines DNA biorepositories with electronic health record (EHR) systems for large scale, high- throughput genetic research in support of implementing genomic medicine. In 2020, eMERGE IV launched a study of genomic risk assessment and return of genomically informed risk assessments (GIRA) with healthcare recommendations for 11 common diseases to a cohort of 23,840 participants (including 4,255 children) from primary care practices across 10 institutions. The primary outcome is analysis of uptake of pre-specified care recommendations (provider action, participant action) across all conditions at 12-months post-GIRA return in high- risk vs. not high-risk participants as defined by the Network-designed statistical analysis plan. Mass General Brigham (MGB) has been an active and engaged eMERGE IV site with specific expertise in EHR data extraction, quality control (QC) processes, and use of large language models (LLMs) to define EHR outcomes. We enrolled and completed GIRA return for 2502 MGB participants, uploaded the results to the EHR, surveyed providers, implemented post- RoR participant surveys, and extracted and QC’d the 6-month EHR outcomes. Extraction of 12- month EHR outcomes is underway and expected to finish in April 2026. Our site has provided Network leadership and was deeply engaged with the other sites and Network leadership throughout the study. In this proposal we request a one-year extension of funding for a “Year 7” from 05/01/2026 to 4/30/2027. This extension will allow the MGB eMERGE IV site to QC and harmonize 12-month outcome data, analyze whether returning high risk GIRA leads to new preventive and screening healthcare actions, and disseminate results through manuscripts to inform implementation of genomic medicine in healthcare for adults and children. We intend to work with the NHGRI leadership and Network sites to continue to discover and publish the learnings from the eMERGE IV genomic risk implementation study following the plans for an extension year that we collaboratively designed as a network.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY Building on our track record in genomic research, clinical trials, and genomic medicine in patients from NYC, we propose to develop new frameworks to bring genomic risk into clinical care to promote preventive health. Polygenic risk scores (PRS) are entering an exciting phase where they are poised to improve health outcomes for myriad complex diseases through enhanced risk stratification and clinical decision making. However, major challenges exist for clinical PRS implementation today, including issues of access to leading-edge genomic technology, research, and testing, and barriers to uptake of medical recommendations. To address this, Mount Sinai experts in statistical genetics and population genetics, with decade-long experience in building methods for genetic risk prediction, will work together to rigorously develop robust clinical PRS tests. We will integrate clinical PRS with standard clinical risk and family history information to generate genomic risk assessments for up to 15 common diseases. Drawing on Mount Sinai’s century of experience delivering excellent patient care, we will recruit 2,500 adult and pediatric patients into a clinical trial. We will estimate participants’ individualized risk for each condition, and investigate the impact of genomic risk communication to patients and their physicians, including patient understanding and uptake of recommended risk-reducing interventions. We will explore attitudes, barriers, and communication preferences related to genomic risk assessment. Knowledge gained will be used to guide the development of a new patient-facing digital platform supporting patient education and communication of genomic risk. We will track patient engagement with their results through the platform, and assess the impact of individualized genomic risk assessments on patient-reported psychosocial outcomes and experiences. As of today, the path to effectively integrate genomic risk into clinical care in busy health systems, is unclear. Hence, we are partnering with clinicians, scientists, industry experts, and community stakeholders to explore a range of strategies to assess, communicate, and reduce disease risk, in order to maximize the efficiency of genomic medicine delivery.

NHGRI - National Human Genome Research Institute

The Baylor College of Medicine node of the NHGRI GREGoR program (BCM-GREGoR) has built on the extensive discoveries and infrastructure established in previous programs. The research program is nested within the Department of Molecular and Human Genetics at BCM and engages the Human Genome Sequencing Center along with other BCM-research activities. Collaborators at the University of Texas and Columbia University Medical Center also participate. Overall, BCM-GREGoR has enrolled a cohort of ~18,000 individuals and families with challenging-to-diagnose rare disease conditions (those unsolved by routine clinical studies such as exome sequencing). Individual cases have been ‘solved’ through the integration of novel methods of genomic data analysis, data sharing across networks, new genomic sequencing technologies, and methods for molecular and organismal phenotypic interrogation of prioritized candidate disease genes and variants. Data have been shared with the GREGoR network via the GREGoR Data Coordinating Center (DCC) and AnVIL, including genomic and phenotypic data, case metadata, and BCM-GREGoR developed genomic tools. This supplement request to support the Baylor College of Medicine GREGoR program will enable completion of the original stated GREGoR goals, including a limited amount of data gathering for participant samples that remain unprocessed and consolidation of all data accrued during the course of the program to support analysis and final conclusions.

NHGRI - National Human Genome Research Institute

Project Abstract This eMERGE IV (E4) supplement proposes extend the outcomes timeline for the prior implementation of 10 “genomic risk assessment” (GIRA) scores in a network-wide set of participants. These GIRAs include polygenic risk score (PRS) information, as well as risk information, such as personal and family health history, environmental and social health determinants, and physical and lab measures. Substantial challenges must be addressed before genomic medicine is a part of standard medical care. The specific aims of our proposal are designed to use an implementation science approach to advance the integration of genomic data into clinical practice, including evaluation of patient perspectives and economic outcomes, and broadening the impact of eMERGE through collaborations. The University of Washington Medicine dedication to preventative health in a learning health system and broad expertise across genomics, statistical, ethical, informatic, implementation, outcomes and economic disciplines will support this multi-site clinical trial. We will continue our leadership in data-cleaning and in the analyses of 6 and 12 month post-return outcomes data.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY/ABSTRACT Genomic sequencing has been rapidly integrated into research and clinical care for common chronic diseases, yet robust evidence remains limited on how genome-informed risk information affects health planning and behaviors and short-term or long-term health outcomes. Returning such information to patients and providers promises to enable earlier, targeted screening, risk-reducing interventions, and tailored disease management. However, if implemented poorly, it can introduce confusion, anxiety, and unnecessary testing. Health systems, providers, payers, and professional societies are being asked to decide which genomic findings to return, how to communicate results, how to check risk updates, and how to support follow- up care—often without real-world outcomes data to guide those decisions. In addition, the rapid pace of advances in genomic analyses presents new translational challenges, which have not been adequately studied in the context of genome informed risks for common chronic diseases that place substantial burdens on individuals, healthcare systems, and society. This proposed research will fill these gaps. As an eMERGE IV site, we have previously developed and validated methods for comprehensive Genome-Informed Risk Assessment (GIRA) for 10 common complex diseases; led risk prediction efforts for two phenotypes (i.e., breast cancer and chronic kidney disease); recruited and returned GIRA risk to 2,536 participants at the Columbia Site; and led the network effort to develop patient- and clinician-facing education, return of results approaches, and ELSI work to consider impacts of GIRAs on patient behaviors and follow up care. We also established standardized and portable clinical decision support infrastructure and data flows in the EHR to support tailored return of GIRA reports. For the 1-year supplement, we proposed to use the time to complete the following activities of the project: (1) Continue the extraction and evaluation of the key study outcome— uptake of pre-specified care recommendations across the ten eMERGE-IV conditions; (2) Assess the impact of GIRA return on multiple condition-specific outcomes, including diagnosis of disease, initiation or intensification of care, and clinical outcomes among high-risk vs. not high-risk participants; and (3) Test the performance of updated genome informed risks compared to legacy risk scores and assess risk status re- classification using in-silico analysis of the eMERGE-IV cohort. Our proposed study will address critical knowledge gaps in the clinical translation of genome-informed risk scores and improve our methods for longitudinal outcome extraction from the EHR and for genetic risk stratification.

NHGRI - National Human Genome Research Institute

NO CHANGE FROM ORIGINAL SUBMISSION Abstract Natural genetic variation impacts most human diseases, yet predicting how regulatory variants control gene expression and ultimately disease phenotypes poses considerable challenges. First, the polygenic inheritance influencing most conditions requires consideration of a vast number of genes and regulatory elements. This task is challenged by the complexity of gene regulation, where 3D regulatory interactions can link enhancers and genes over large genomic distances. Second, multiple interacting cell types are often dysregulated in disease pathology. This necessitates an understanding of how the collective variants associating with a disease affect each cell type involved in the disease process and subsequently how these dysregulated cellular phenotypes crossregulate and drive subsequent cellular states. In this IGVF project, we will use rheumatoid arthritis (RA), a human autoimmune inflammatory disease, as a case study to develop robust machine learning models of gene regulation to decipher the impact of genomic variation on multiple cellular drivers of pathology—namely, inflammatory T cell and fibroblast subsets found in affected joint tissue. The choice of RA is motivated by its public health importance, specified target tissue, access to clinical samples, considerable knowledge of disease-associated gene loci, and our team’s complementary expertise in machine learning, RA pathophysiology, immunology and inflammation, and single-cell functional genomics. We will develop an advanced machine learning framework to model the effects of allelic variation on gene regulatory networks based on the analysis of epigenomes, transcriptomes, and connectomes of mouse activated T cells and synovial fibroblasts and extend these models to RA patient joint tissue and primary cells. We will train allele-specific gene regulatory models (GRMs) that account for long-range regulatory interactions by integrating single-cell transcriptome and epigenome (sc-multiome) data with bulk 3D interactome analyses. A notable feature of our approach is that we leverage the genetic diversity of evolutionarily distant F1 hybrid mice to provide robust training data for these models, and then apply these advances to the human context through transfer learning. Highly parallelized Perturb-seq experiments in primary synovial fibroblasts from RA patients with single-cell multiomic readouts will then be used to evaluate and refine regulatory models and to train network models that connect gene expression programs to phenotype. Finally, we will combine spatial and single-cell transcriptomics conducted on samples from RA inflamed joints to model the organization and interactions between T cells and sedentary tissue-organizing fibroblasts within local cellular communities. The predictive GRMs that will be generated from our study along with the experimental systems for human disease will be readily transferrable to other polygenic disorders which must consider complex regulatory genomic networks for various interacting cell types in affected tissues.

NHGRI - National Human Genome Research Institute

Project Summary/Abstract Large-scale studies have identified thousands of genetic variants linked to developmental defects, together with the regulatory elements harboring these variants and the cell types in which these variants likely function. This diversity of variants, regulatory elements, and cell types indicates that multiple mechanisms contribute to developmental defects. One key challenge to our understanding of these mechanisms is that the molecular, cellular, and functional phenotypes of each variant remain largely uncharacterized. Until these critical gaps in knowledge are addressed, the underlying molecular and cellular determinants of developmental disease susceptibility will remain incomplete. To bridge these gaps, we propose to establish the “UT Southwestern Center for Regulatory Element Variation and Function”. The primary goal of this Center is to systematically catalog molecular and cellular phenotypes for disease-associated enhancers in human development, with a focus on gaining insights into mechanisms of non-canonical human genetics and gene regulation. To build a generalizable framework to understanding the impact of human genetic variation on function, we propose a high throughput perturbation platform with three primary goals: (1) Contribute to a variant/element/phenotype catalog with relevance to diseases of human development, focusing on elements genetically associated with congenital heart disease (cardiomyocytes), autism (neurons), and placental defects (trophoblasts); (2) Contribute to a variant/element/phenotype catalog for non-canonical human genetics, focusing on two understudied topics in human genetics: pleiotropic effects and non-cell autonomous effects; and (3) Contribute to a variant/element/phenotype catalog with relevance to mechanisms of gene regulation, focusing on enhancer RNAs. The Center will take advantage of recent technological innovations in genome engineering, single-cell genomics, and high content screening to enable the multiscale functional characterization of genomic variation in human developmental disorders. Several of these techniques have been pioneered by investigators contributing to this project, including: the development of novel tools for enhancer perturbation and the coupling of endogenous enhancer perturbations with a single-cell RNA-Seq readout (Mosaic-Seq). Impact and Significance: The efforts on this project will lead to a number of key outcomes and deliverables, including (1) greater understanding of the relationships between sequence variation and genome function, (2) an extensive variant/element/phenotype catalog for the community, (3) tools for generating predictive models for the community, and (4) resources to enable future functional genomics studies. Together, our multifaceted and combinatorial approaches will open new horizons to understanding the impact of regulatory variants on developmental disease phenotypes.

NHGRI - National Human Genome Research Institute

ABSTRACT Noncoding genetic variation that alters gene regulation is of paramount importance for health, disease, and evolution. Diseases ranging in incidence from the most common to the most rare all have substantial risk associated with regulatory variation; and most of the genetic differences between closely related species are noncoding. Whole genome sequencing can directly identify that variation but to realize its potential to elucidate the genetic determinants of health and disease, will require accurate annotation of this noncoding variation for functionality. In coding sequence, the genetic code allows variants to be annotated to a rough hierarchy of likely functional effects and pathogenicity. In noncoding sequence such annotation is less clear. Perturbation assays, i.e., assays that modify genetic or epigenetic states and measure the effect of those perturbations on regulatory endpoints, offer a possible path to annotating noncoding variation. However, to fully leverage this data, novel and sophisticated statistical and machine learning approaches are required to extract useful information from those assays, to integrate that information across regulatory endpoints, and to extrapolate findings so that annotation of previously unobserved (unperturbed) variation in diverse cell types is possible. The goal of the Duke Prediction Center is to develop the analytic approaches and tools that will allow for the routine annotation of noncoding variation for functionality and ultimately pathogenicity. Aim 1 is to establish best practices in perturbation assay design and analysis. This will allow IGVF characterization centers design their experiments so that, when coupled with optimized analyses, the data produced will be maximally informative for subsequent predictive modeling. Aim 2 is to develop novel mechanistic machine learning approaches for predicting the functional effect of noncoding variation on function in diverse cell-types. Aim 3 is to identify noncoding genomic regions that are subject to functional constraint which will be leveraged in prioritizing variants for pathogenicity. The expected outcomes of this project will be (i) robust estimates of optimal experimental design parameters and recommendations for analysis tools and best practices for the various assays used within the IGVF consortium, (ii) predicted functional effects of observed variation to be shared through the IGVF variant/phenotype catalog as well as a state-of-the-art machine learning method (and associated tools) that can identify previously-unknown interactions among genomic variants, both observed and novel, and predict their functional impact in diverse cell types, and (iii) a list of regulatory elements subject to functional constraint shared through the IGVF variant/phenotype catalog and a principled prioritization framework (and associated tools) for interpreting variation within patient genomes for pathogenicity. Due to the considerable success of genetics, there are thousands of unknown regulatory causes of disease. Each of those causes is an opportunity to improve treatment, diagnostics, or prevention. This project will be a major advance towards unlocking that potential.

NHGRI - National Human Genome Research Institute

ABSTRACT Large scale genome annotation consortia such as ENCODE, Epigenomics Roadmap, and others have identified millions of putative regulatory elements. We now need to focus efforts on comprehensively characterizing and quantifying the function of those elements, and noncoding variants that map within these regions, on gene expression and cell phenotypes. Our long-term goal is to assign function to every regulatory element and noncoding variant in the human genome, understand how that function changes in different contexts, and use that information to better understand cell fitness, disease mechanisms, cell lineage specification, and tissue homeostasis. To accomplish this goal, we have developed multiple novel high-throughput CRISPR-based technologies for characterizing the function of putative gene regulatory elements by perturbing their activity in their endogenous, native context. We have coupled these methods with single-cell RNA-seq to identify the target gene(s) for each regulatory element. We have also developed dCas9 effector mice to characterize elements in their natural in vivo context. In addition, we have developed population-based high-throughput reporter assays (POP-STARR) to characterize the impact of noncoding genetic variation across the entire genome. The objective of this proposal is to apply and share our compendium of complementary, robust, scaleable, and well-characterized methods by working collaboratively to support the IGVF Consortium goals of understanding how genomes and genomic variation function and orchestrate complex phenotypes. Our track record in developing, applying, and sharing these high-throughput characterization methods, as well as providing access to all data, supports that we will be successful in accomplishing our objective via the following specific aims: Aim 1. Characterize all gene regulatory elements essential for cell survival. Aim 2. Characterize all gene regulatory elements essential to cell lineage specification. Aim 3. Characterize all gene regulatory elements in select eQTL regions. Aim 4. Characterize all non- coding elements essential to tissue homeostasis in a mouse model. We will make all data immediately available, as well as share comprehensive protocols, reagents, and analysis tools to the scientific community. Together, the diverse approaches of this Characterization Center will lead to transformative progress in understanding the role of regulatory elements and noncoding variants across many diverse phenotypes.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY/ABSTRACT A comprehensive genome-wide map of DNA regulatory elements and gene expression in human cells is of critical importance for understanding how genomic variation impacts human health and disease. Since regulatory DNA elements are exceptionally cell type-, tissue-, and disease state-specific, a comprehensive catalog of these elements has been difficult to achieve. The overall mission of this IGVF Mapping Center is to create a high quality, open-access, and single cell-resolution reference map of human regulatory elements and gene expression in immune cells during human development, across organ systems in healthy adults, and in tissues from diverse immune-related diseases. Our Mapping Center will leverage: (i) our recent advances in developing scalable and cost-efficient single-cell epigenome and multi-omic technologies to simultaneously map open chromatin sites, gene expression, intracellular and cell surface proteins, and clonal lineage tracing in each tissue sample, (ii) our prior technical improvements and application of these methods to primary tissues from humans, and (iii) our pre-existing human tissue biobank consisting of samples from more than 500 human individuals, 20 organ systems, and 15 disease conditions, consented for unrestricted access, genomic sequencing, and data sharing. In Specific Aim 1, we will work closely with the IGVF Consortium to establish cross-center plans for data generation, analyses, and effective coordination of sample access and sharing. In Specific Aim 2, we will generate a single-cell multi-omic atlas of immune cell types (and non-immune cells types, as determined with the IGVF) during development in early life, healthy aging, and across human organ systems. In Specific Aim 3, we will generate a single-cell multi-omic atlas in immune cell types from primary tissues in patients with autoimmunity, cancer, neurodegenerative disease, and infection. In Specific Aim 4, we will analyze regulatory sites and gene expression in the context of clonal differentiation trajectories inferred from mitochondrial lineage tracing and develop and maintain an integrated reference map of each datatype and tissue sample for the research community. Our Mapping Center, composed of 7 new investigators with extensive experience in single cell genomic technologies and human disease analysis, will work closely with the IGVF to share technologies, resources, data, and tissue samples towards the shared goal of developing a comprehensive single-cell atlas of cell types, functional regulatory elements, and gene expression in humans.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY/ABSTRACT Sequencing human genomes, especially those from individuals with Mendelian disorders, has yielded discovery of thousands of new disease genes and has led to personalized treatments, improving the lives of patients and their families. Unfortunately, less than half of this population is able to reap these benefits because, despite continued advances in sequencing technology and data analysis, many patients remain without a molecular diagnosis. A major barrier to increasing the diagnostic yield is the abundance of variants of uncertain significance (VUS), and in particular missense VUS, which are found in more than half of all patients. As more individuals are sequenced, the number and proportion of patients with missense VUS continues to grow, much faster than the field is currently able to interpret the functional significance of these variants. Therefore, it is critical to both scale up existing methods and develop new approaches to determine the pathogenicity of these VUS in as many genes as possible to help patients and families in need. One such method that offers hope for solving this VUS problem is Deep Mutational Scanning (DMS), which uses cell-based assays to simultaneously test thousands of missense variants by changing each residue in a protein of interest to all 19 other possible amino acids. However, it is currently very challenging to generate a cell-based assay that reliably reports the activity of a given gene of interest. Fortunately, combining DMS with a multiplexed fluorescence-based approach, SortSeq, gives an accurate assessment of deleteriousness and pathogenicity for variants in the GLI2 gene, which operates in the Sonic Hedgehog (SHH) signaling pathway. This proposed work will use this same well-validated assay to interrogate multiple genes involved in SHH signaling in order to determine the feasibility of a pathway-based approach for efficiently conducting DMS in many genes with the same molecular read-out. The scope of work is expanded to include a second pathway, the retinal development pathway, including DMS of CRX and OTX2, and leverages identical assays for paralogous families of transcription factors to further scale up the approach. This proposal also tests and validates an innovative computational pipeline that includes de novo biophysical predictions to understand the effects of variants in intrinsically disordered regions (IDRs), which is particularly synergistic with the focus on transcription factors, which typically contain large stretches of IDRs. Successful completion of these aims will establish multiple strategies for pathway-based and paralog-based scaling that can be quickly and easily adapted across many disease-causing genes, resulting in improved genome-based diagnostics and future therapeutics.

NHGRI - National Human Genome Research Institute

The ML/AI Tools to Advance Genomic Translational Research (MAGen) consortium is a national, highly functional, network infrastructure that will enhance the accuracy and precision of predicting how individuals with pathogenic variants manifest disease. Our MAGen Coordinating Center (CC) Team is composed of leading experts at Vanderbilt University Medical Center. As the CC, we will support the Development Sites (DS) to synthesize Genetic variant Evidence using Novel tools to Elucidate Pathophysiology: Accelerating Translation to Health’ (called ‘GENEPATH’) as well as support the Consortium in defining critical connections to optimally create tools that provide a holistic prediction with explanation of how a variant causes disease in an individual in the context of their life along with exploring the ethical, legal, and social implications (ELSI) of integrating ML/AI tools into genomic medicine. This work requires that GENEPATH bring together experts on variant interpretation, protein function, genomic medicine, genetic anthropology, informatics, genomic consortium coordination, and all aspects of ELSI, along with our unique skills in consensus building to support DSs in coordination, ELSI research projects, and development of a common data model. Functionally, 1) we will serve as a central home for the Consortium by implementing the Scientific Operations Unit. This Unit will coordinate all Consortium activities including supporting the Consortium and its Steering Committee to establish, monitor, and reach program goals, providing project management with deep knowledge of machine learning, ELSI, and genomic medicine to ensure milestones are met, and structuring Consortium collaboration to promote synergy. 2) We will establish a flexible technical architecture adoptable by development sites through the Data & Machine Learning Unit which will create common data models that handle multi-domain, structured and unstructured data, and plan the cross-validation protocols, as well as collaboratively define specifications for AI/ML variant characterization tools. 3) We will build trust and credibility through the Engagement in ELSI Unit by ensuring authentic communication with patients, communities, and providers to guide Consortium planning. GENEPATH efforts will allow us to broaden our understanding of how variants manifest in disease, leading to a more precise and effective use of genetic variation in research and healthcare.

NHGRI - National Human Genome Research Institute

ABSTRACT DNA methylation (DNAme) is a core layer of epigenetic regulation with central roles in the establishment and maintenance of cellular identity. Methods to profile DNAme at single-cell resolution are needed to elucidate epigenetic networks governing cell state in healthy tissues and understand their dysregulation in disease and aging. However, existing methods for single-cell methylome profiling are highly inaccessible, and no methods to experimentally profile both DNAme and spatial location currently exist. I have developed a novel method for single-cell DNAme profiling that leverages the widely available 10x Single Cell Multiome and NEB EM-seq kits, which I have named droplet-implemented single-cell DNA methylation sequencing (discDNAme). Applied to nuclei isolated from mouse brain, discDNAme recovered ~3,000 high-quality methylomes that clustered into clearly separated neuronal and non-neuronal subtypes and displayed stereotyped patterns of CpG dinucleotide methylation around key genomic features. However, these measurements lack spatial information on the native tissue contexts of profiled cells, and the protocol’s per-cell library size is lower than current gold-standard methods for measuring single-cell DNAme. I propose (1) development of a spatially resolved single-cell DNA methylation technology by combining discDNAme with slide-tags, a platform for spatially resolved single-nucleus RNA- and/or ATAC-seq developed by our group. To integrate slide-tags with discDNAme, I will develop a protocol in which the “spatial barcode oligos” we use to position nuclei are physically separated from genomic DNA prior to unmethylated cytosine conversion, benchmark this technology in the mouse hippocampus, and apply it to study glioblastoma multiforme. I further propose (2) experiments to improve and benchmark my discDNAme technology. I will systematically test various independent approaches to increase library complexity at different steps of the discDNAme protocol, combine these optimizations into a second-generation protocol, and benchmark this against our original protocol and other leading methods for single-cell DNAme profiling. Completion of this proposal will result in (1) the first high-resolution method to measure spatially resolved single- cell methylomes—a major advancement in spatial omics technologies—and (2) an accessible yet capable tool for single-cell DNAme profiling that will open single-cell DNAme studies to the broader single-cell community.

NHGRI - National Human Genome Research Institute

Project Summary Polygenic risk scores (PRS) are becoming increasingly accessible through direct-to-consumer (DTC) platforms and clinical practice. While PRS have the potential to enhance health monitoring and motivate preventive behaviors, misinterpretation of probabilistic risk estimates may also induce distress, reinforce genetic determinism, and lead to misguided health-related decisions. Present research indicates that patients and the public frequently misunderstand PRS results, often overestimating genetic influence and failing to contextualize probabilistic risk. Additionally, despite the rapid expansion of PRS use, standardized guidelines for communicating their meaning, limitations, and predictive accuracy remain lacking. This mixed-methods research program aims to systematically examine PRS comprehension, identify psychological predictors of genetic determinism, and develop evidence-based educational interventions to enhance genetic risk communication and comprehension. Aim 1 will assess how individuals understand genetic and statistical concepts underlying PRS through the development and evaluation of infographics designed to improve comprehension. Additionally, this aim will investigate how genetic causal attributions and genetic beliefs shape PRS interpretations, psychosocial responses, and health decisions. Aim 2 will experimentally evaluate how different PRS result formats (percentile rank vs. percentage change) and the inclusion of predictive accuracy metrics influence risk perception, emotional reactions, and subsequent health behaviors. Aim 3 will build upon these findings by experimentally testing PRS infographics integrated with a comprehension assessment and targeted corrective feedback to enhance public understanding and informed health decision-making. Overall, this research will identify key comprehension barriers, determine effective communication strategies, and establish best practices for responsibly reporting PRS results in both clinical and DTC settings. Findings will provide critical insights for healthcare providers, policymakers, and genomic companies to optimize PRS integration into precision medicine, empowering individuals to make informed, personalized health choices. This research will further serve as the foundation for a future R01-funded longitudinal randomized controlled trial (RCT) evaluating the long-term effectiveness of PRS communication strategies on psychosocial and behavioral health outcomes. By improving PRS comprehension, this work will help bridge the gap between scientific advancements in genomics and their ethical, effective application in public health.

NHGRI - National Human Genome Research Institute

Project Summary/Abstract The rapid advancement of genomic medicine holds great promise for improving diagnosis, treatment, and health outcomes across diverse populations. However, the pace of progress has outstripped the ability of many healthcare providers to effectively integrate this knowledge into their practice, a challenge compounded by a limited genetic workforce. As a result, a widening gap has emerged between the potential of genomics and its practical application, particularly in underserved communities where access to genetic expertise is severely constrained. To bridge this divide, electronic consultations (eConsults) have emerged as a promising solution, enabling timely and cost-effective communication between referring providers and genetic specialists, thereby expanding access to critical expertise and support. The University of California, San Francisco (UCSF) proposes to address these gaps in care delivery by establishing the Western Regional Genomic eConsult Network (WestGEN), a comprehensive genomic medicine eConsult service for the Western United States available to both primary care providers (PCPs) and specialists, with an emphasis on supporting underserved populations. WestGEN will conduct studies to support the integration of genetic counselors into the eConsult service, and pilot AI solutions to augment genomic eConsults. The program incorporates research to optimize the delivery of genomic eConsult services and assess its adoption and impact. The specific aims are: 1) Deploy WestGEN across the Western U.S., initially to 239 primary care practices through an existing eConsult platform, and then extend to eConsult-naive practices; establish the role and scope of practice for genetic counselors; and evaluate genomic eConsult utilization and outcomes using mixed methods. 2) Support the expansion of regional genomic medicine eConsult programs nationwide by developing implementation tools and best practices for sustainability. 3) Evaluate an AI-assisted eConsult model to enhance efficiency and scalability. WestGEN brings together experts in genomics, implementation science, eConsults, and machine learning to create a scalable, evidence-based model for genomic medicine eConsults. The project will provide clear evidence on the impact of genomic eConsults on patient outcomes, generate tools for dissemination, and explore an AI-assisted model to improve efficiency and reach. By achieving its aims, WestGEN will enhance access to genomic expertise, provider knowledge, and quality of care, particularly for underserved populations, aligning with NHGRI's mission to implement genomics in clinical care to improve health equity and outcomes for all.

NHGRI - National Human Genome Research Institute

PROJECT SUMMARY Translating knowledge of the genetic basis of disease into better health outcomes is the major promise of translational genomic medicine and precision medicine. However, common, actionable genomic conditions with well-understood genetic components and effective early detection and treatment protocols, such as hereditary breast and ovarian cancer, Lynch syndrome, and familial hypercholesterolemia, are too often not detected by current screening protocols. Population-based genomic screening for all adults in primary care settings holds the promise of earlier identification of those at risk, leading to reduced morbidity and mortality through earlier monitoring, treatment, and other interventions. The Population Genomic Screening Network (PGSN) will pilot genomic screening in primary care settings for these and select additional genomic conditions. PGSN will recruit over 20,000 adult participants across diverse clinical settings, targeting health disparities populations. As the PGSN Coordinating Center, we will (1) coordinate and monitor the Network’s design and implementation of the screening program, including serving as the single IRB, coordinating consensus-driven development of a screening program implementation manual, monitoring and reporting on screening program implementation processes, and supporting the Network’s community engagement plan. We will also (2) manage and release the Network’s data, including coordinating and monitoring the secure transfer and management of data across the Network, developing a data model to support data aggregation and harmonization, and producing a cleaned, well-documented, de-identified Network dataset that will be released to the scientific community. We will also (3) contribute to the collaborative development and dissemination of best practice and lessons learned from the pilot program. Lastly, we will (4) provide comprehensive administration of PGSN’s logistics, communication, and governance including scheduling and managing all Network calls, organizing in-person/hybrid meetings, overseeing communications and calendaring, and hosting a comprehensive website.