Notice of Special Interest: Development of Statistical, Population Genetics and Computational Methods Related to Polygenic Prediction of Health and Disease in Diverse Populations NOT-HG-20-010

Key Dates

Related Announcements

RFA-HG-20-001, Polygenic Risk Score (PRS) Methods and Analysis for Populations of Diverse Ancestry Study Sites (U01 Clinical Trial Not Allowed)

RFA-HG-20-002, Polygenic Risk Score (PRS) Methods and Analysis for Populations of Diverse Ancestry Coordinating Center (U01 Clinical Trial Not Allowed)

PA-19-056: Research Project Grant (Parent R01 Clinical Trial Not Allowed)

PA-19-053: NIH Exploratory/Developmental Research Grant Program (Parent R21 Clinical Trial Not Allowed)

Issued by

National Human Genome Research Institute (NHGRI)

National Heart, Lung, and Blood Institute (NHLBI)

National Cancer Institute (NCI)

Purpose

Purpose

The purpose of this Notice of Special Interest (NOSI) is to solicit investigator-initiated applications to develop statistical, computational, and population genetics approaches to modeling polygenic risk in human populations, especially in genetically diverse populations.

Polygenic risk scores (PRS) developed using large-scale genomic data from epidemiological studies are rapidly becoming linked to clinical implications, such as identifying individuals who would particularly benefit from modification of coronary heart disease risk factors. Currently available scores, however, show poorer risk prediction in non-European populations due to the vast under-representation of non-European ancestry (EA) populations in the underlying GWAS data. Methods that factor in population diversity and are applicable across populations are needed. Additionally, the choice of statistical models used to predict PRS, underlying assumptions about allelic architecture and population stratification, and the application of the models to predict health and disease risk are heterogeneous across research studies. This heterogeneity may limit the ability to validate and compare PRS across studies and motivates the need to develop methods, standards, and approaches to studying PRS in large-scale studies of diverse populations.

In May 2019, NHGRI and NHLBI convened two multidisciplinary workshops, Genomic Medicine XII: Genomics and Risk Prediction and Polygenic Risk Scores: From Discovery to Implementation. A prominent recommendation from both meetings was support of research, including methods development, to understand PRS and its application in diverse, non-European populations to adequately capture genetic variation and improve methods to integrate traditional and genetic risk factors for more accurate estimation of participant disease risk.

As a follow-up to these meetings, NHGRI recently proposed an initiative for collaborative data integration and analysis of PRS from populations of diverse ancestry to the National Advisory Council for Human Genome Research (https://www.genome.gov/sites/default/files/media/files/2019-09/Sept2019_Council_Hindorff_PolygenicRiskScore_RFA_Concept.pdf). As a result of that discussion (https://www.youtube.com/watch?v=x0-hw7DoOQU), NHGRI identified a need for methods development related to modeling of polygenic risk of health and disease in genetically diverse populations. There is currently a lack of consensus on approaches to generating PRS, especially in genetically diverse populations. The incorporation of the degree of genetic ancestry or admixture, the predicted functional impact of genetic variants, and other health-related data may predict additional risk beyond the genetic variant information alone. This NOSI encourages investigation into novel -omic and ancestry-informed methods that may result in more nimble and creative approaches to risk prediction and data integration. Development of methods that can be used broadly and that will leverage polygenic prediction of health and disease to understand foundational questions related to genomics and biology of disease are particularly encouraged. Existing large-scale datasets, such as those from the Trans-Omics Precision Medicine (TOPMed) Program (https://www.nhlbiwgs.org/) and Centers for Common Disease Genomics (https://ccdg.rutgers.edu/), have been, and could continue to be, utilized for methods development. To broaden the reach of the methods development work funded through this NOSI, investigators funded by this NOSI are encouraged to develop robust plans for data sharing and may be invited to participate in related consortia or cooperative agreements, either existing or to be funded.

Topics of Interest

Topics to be addressed by this NOSI include novel methods to improve prediction of polygenic influence on health and disease, with an emphasis on ancestrally diverse populations. Applications in response to this NOSI should address one or more of the following priority areas:

• Statistics or epidemiology
  • Integrating multidimensional phenotype information into PRS
  • Integrating clinical information into PRS
  • Addressing PRS heterogeneity and prediction accuracy across populations, especially among under-represented, non-European populations.
  • Modeling differences in risk prediction across subgroups, such as those defined by genetic diversity, disease subtypes or sex, with an emphasis on non-EA populations
  • Integrating PRS with existing risk stratification models to identify individuals at particularly high or low risk of disease, or at the extremes of health status
  • Evaluating the performance of different PRS models on the ability to predict polygenic influence on health and disease
• Population genetics
  • Incorporating diverse genetic ancestry or admixture in models of polygenic risk
  • Modeling polygenic risk in admixed populations
  • Using population genetics to study why PRS differ across populations and how to account for these differences
• Computational or functional genomics
  • Incorporating functional information about genetic variants into models of polygenic risk
  • Developing network or pathway approaches to integrate information on genotype, phenotype, and other -omics to predict polygenic risk
  • Developing methods to deconvolute the black box of PRS scores and improve understanding of variant function

Applications to this NOSI will focus primarily on methods development and not on the application of PRS to large-scale clinical data or clinical implementation of PRS. No genomic data generation or results disclosure to participants will be funded through this NOSI.

NCI Priorities

NCI is interested in PRS methods development studies that focus on cancer.

NHLBI Priorities

NHLBI is interested in PRS methods development studies that focus on Heart, Lung, Blood or Sleep (HLBS) disorders or established risk factors for HLBS disorders.

Application and Submission Information

This notice applies to due dates on or after February 5, 2020 and subsequent receipt dates through October 5, 2023.

Submit applications for this initiative using one of the following funding opportunity announcements (FOAs) or any reissues of these announcements through October 5, 2023.

• PA-19-056: Research Project Grant (Parent R01 Clinical Trial Not Allowed) [NHGRI, NCI AND NHLBI applications].
• PA-19-053: NIH Exploratory/Developmental Research Grant Program (Parent R21 Clinical Trial Not Allowed) [NHGRI applications]?.

All instructions in the SF424 (R&R) Application Guide and the funding opportunity announcement used for submission must be followed, with the following additions:

For funding consideration, applicants must include NOT-HG-20-010 (without quotation marks) in the Agency Routing Identifier field (box 4B) of the SF424 R&R form. Applications without this information in box 4B will not be considered for this initiative.

All applicants are strongly encouraged to contact NIH Staff (see Scientific/Research Contacts) to discuss the alignment of their proposed work with the goals of this FOA and the scientific mission of the respective NIH Institutes.

Applications nonresponsive to terms of this NOSI will be not be considered for the NOSI initiative.

Inquiries

Please direct all inquiries to:

Scientific/Research Contacts:

Joannella Morales, Ph.D.
Program Director, Division of Genomic Medicine, NHGRI
Telephone: 301-529-7477
[email protected]

Leah Mechanic, PhD, MPH
Program Director, Genomic Epidemiology Branch, Epidemiology and Genomics Research Program, NCI
Telephone: 240-276-6847
[email protected]

Cashell Jaquish, PhD
Program Director, Division of Cardiovascular Sciences, NHLBI
Telephone: 301-435-0707
[email protected]