NHGRI logo

Frequently Asked Questions about RFA-HG-07-014

Epidemiologic Investigation of Putative Causal Genetic Variants - Study Investigators (U01)

October 9, 2007
Will genotyping be done at NHGRI, or should each investigator propose genotyping? Will genotyping be done at NHGRI, or should each investigator propose genotyping?
What are considered reasonable costs for GWA genotyping? What are considered reasonable costs for GWA genotyping?
For study populations that were included in a GWAS and already have genome-wide genotyping, how would this funding be used? For study populations that were included in a GWAS and already have genome-wide genotyping, how would this funding be used?
Would funding for functional studies be considered? Would funding for functional studies be considered?

This RFA appears to be an investigation of only variants previously identified by GWA that have already been independently replicated; is this correct? This RFA appears to be an investigation of only variants previously identified by GWA that have already been independently replicated; is this correct?
The RFA calls for population-based studies.  Are community-based family studies acceptable? The RFA calls for population-based studies. Are community-based family studies acceptable?
Many potential causal variants have been identified in family studies through quantitative trait analyses (linkage and association). Can variants identified in this way be proposed for follow-up, or it is just variants identified from GWAS? Many potential causal variants have been identified in family studies through quantitative trait analyses (linkage and association). Can variants identified in this way be proposed for follow-up, or it is just variants identified from GWAS?
Can a study with only case-control data available be used? Can a study with only case-control data available be used?

Would studies with extensive cross-sectional, but not longitudinal, measures of demographics, lifestyle, lab measures, etc. be responsive, especially if longitudinal data will become available during the course of the grant? Would studies with extensive cross-sectional, but not longitudinal, measures of demographics, lifestyle, lab measures, etc. be responsive, especially if longitudinal data will become available during the course of the grant?
Will cohort studies in which enrollment has not been completed, nor will not start until partway into the funding period of the grant be acceptable? Will cohort studies in which enrollment has not been completed, nor will not start until partway into the funding period of the grant be acceptable?
May cohorts in which stored DNA is available on only a small fraction of participants be proposed for study under this RFA? May cohorts in which stored DNA is available on only a small fraction of participants be proposed for study under this RFA?
Would a Northern European cohort be considered relevant to the US population? Would a Northern European cohort be considered relevant to the US population?

Is it expected that a proposal will cover multiple (perhaps unrelated) disease areas, or would focusing on certain genes and their relationship to a disease and related traits suffice? Is it expected that a proposal will cover multiple (perhaps unrelated) disease areas, or would focusing on certain genes and their relationship to a disease and related traits suffice?
Would it be possible to combine smaller cohorts to meet the size requirements of the RFA? Would it be possible to combine smaller cohorts to meet the size requirements of the RFA?
Can we apply to this program with linked R01s or identify more than one Principal Investigator? Can we apply to this program with linked R01s or identify more than one Principal Investigator?
Does the U01 mechanism have different requirements than the R01 mechanism? Does the U01 mechanism have different requirements than the R01 mechanism?
What are the data sharing requirements for this research program? What are the data sharing requirements for this research program?

Will genotyping be done at NHGRI, or should each investigator propose genotyping?

Genotyping costs will not be covered by a separately funded genotyping center (either at NIH or elsewhere). Each applicant should propose a genotyping plan and corresponding budget, as described in Section 1, "Funding Opportunity Description," of the RFA: "Genotyping will be performed by the awardees under this RFA, not under a separately-funded genotyping center, and costs for this activity should be proposed by the applicant." Note that this RFA will not support genome-wide association (GWA) genotyping.

What are considered reasonable costs for GWA genotyping?

To clarify - this RFA will not support genome-wide association genotyping, defined as genotyping of hundreds of thousands of single nucleotide polymorphisms or other genetic markers designed to capture the majority of human genomic variation. The goal of this research program is to genotype putative causal variants which have been well-replicated in a number of genetic studies (including initial GWA discovery and replication studies, as well as linkage, candidate gene, or other study designs) in existing population-based cohort studies and clinical trials to examine their prevalence, correlations, and potential modifiers on a population basis. Each applicant should propose and justify an approach for identifying and rapidly genotyping these variants, relating these individual genotypes to known characteristics of the study subjects, and disseminating the findings from these analyses. Costs will vary with the number of variants proposed, types of samples in hand, etc. The costs proposed should be the best research costs negotiable with vendors and laboratories. As with all cost estimates, they will need to be well justified and documented.

For study populations that were included in a GWAS and already have genome-wide genotyping, how would this funding be used?

Note that this RFA will not support genome-wide association (GWA) genotyping; genotyping will be supported only for a selected number of putative causal variants. If GWA genotyping has already been done and putative causal variants proposed for study have already been genotyped, applicants will be able to proceed directly to analysis of prevalence and correlations, though it might be expected that the initial GWA genotyping support might also have covered these analyses. Applicants might consider emphasizing to peer- reviewers the advantages of available genome-wide genotyping data on the proposed study population, and also making it clear that there will be no duplication of funding or effort in previously supported studies. In addition, it is likely that additional SNPs surrounding the proposed variants of interest will need to be typed; studies with existing GWA data may be able to type a larger number of such variants.

Would funding for functional studies be considered? (e.g. if gene found via GWAS, could we add a component of zebrafish, for example?)

No, the focus of this RFA is performing genotyping and related assays in the available participant samples. Funding will not be provided for functional studies in model organisms. Applicants who already have collaborations set up for future functional research may provide a more attractive setting in which to carry out the goals of this RFA, so applicants may want to describe these additional collaborations.

This RFA appears to be an investigation of only variants previously identified by GWA that have already been independently replicated; is this correct?

Not exactly; this RFA provides an opportunity to better characterize genetic variants that were previously identified via genome-wide association studies or other types of genetic studies with the hope of understanding their "population impact" by evaluating the variants in well-characterized, large cohorts or trials. Priority will likely be given to variants with strong evidence suggesting they are causally associated with a disease or trait. This RFA is not designed to provide a standing resource for rapid replication of GWAS "hits" for publication, although it is intended that detailed descriptions of the funded studies will be posted in dbGaP and on the program website. This will allow investigators who need another population sample for replication to easily identify which cohorts/large trials might be available for collaboration. Such collaborations will be facilitated, but not supported, by this RFA.

The RFA calls for population-based studies. Are community-based family studies acceptable?

For the purposes of this RFA, "population-based" means representative of and generalizable to the U.S. community-dwelling population. Large, population-based cohort studies or trials with detailed information on demographics, health characteristics, environmental exposures, and disease risk factors and traits are the primary focus of this RFA. Applicants proposing to use family-based studies will need to demonstrate to peer reviewers that such studies can appropriately serve the goals of this RFA and findings can be generalized to the U.S. population.

Many potential causal variants have been identified in family studies through quantitative trait analyses (linkage and association). Can variants identified in this way be proposed for follow-up, or it is just variants identified from GWAS?

Yes, variants identified in family studies would be appropriate. For the purposes of this RFA, "putative causal variants" are defined as genetic variants for which there is strong statistical, biological, and/or functional evidence of association with a disease or trait such that the variant is likely to play a role in etiology. Any variant, as long as it is well justified, would be considered. Note that criteria for selecting variants for follow-up will be developed within this collaborative program, by the program Steering Committee (PIs from each investigative group, the Coordinating Center, and the NIH project scientist).

Can a study with only case-control data available be used?

No. This RFA is designed to utilize existing large, well-characterized population-based cohort studies and clinical trials.

Would studies with extensive cross-sectional, but not longitudinal, measures of demographics, lifestyle, lab measures, etc be responsive, especially if longitudinal data will become available during the course of the grant?

Yes, as noted in the RFA,"...existing information may be cross-sectional, if extensive, or, preferably, prospective and longitudinal." Given the limitations of cross-sectional data, applicants proposing such studies will need to demonstrate to peer reviewers that such studies can appropriately serve the goals of this RFA, and may wish to emphasize any unique or compelling strengths of the datasets they are proposing.

Will cohort studies in which enrollment has not been completed, nor will not start until partway into the funding period of the grant be acceptable?

Although applications proposing such studies would be considered broadly responsive to the RFA, they may be compared to other applications that already have well-characterized longitudinal data and DNA samples available for genotyping. Applicants will need to convince peer reviewers of the unique and compelling strengths of such a study and that it can meet the RFA goals. The RFA will not support de novo sample or data collection on study participants.

May cohorts in which stored DNA is available on only a small fraction of participants be proposed for study under this RFA?

No. Samples are expected to have been collected by the time of funding. The RFA will not support de novo sample or data collection on study participants, though in extraordinary cases consideration may be given to repeat collection in participants whose samples have been exhausted in prior studies.

Would a Northern European cohort be considered relevant to the U.S. population?

Possibly; for this particular research program, studies involving U.S. racial or ethnic minorities, particularly those experiencing disproportionate burdens of disease, will receive higher programmatic priority. Racial and ethnic makeup of a northern European population proposed for study should be detailed and its relevance to the US population demonstrated. Applicants proposing such studies will need to demonstrate to peer reviewers that such studies can appropriately serve the goals of this RFA, and may wish to emphasize any unique or compelling strengths of the datasets they are proposing.

Is it expected that a proposal will cover multiple (perhaps unrelated) disease areas, or would focusing on certain genes and their relationship to a disease and related traits suffice?

The RFA calls for studies that are of sufficient size and breadth to permit examination of a wide range of variants and traits, and to identify associations and interactions with sufficient power. Applicants may propose a set of hypotheses and analyses based on variants related to a particular disease and closely related traits, but should also show that there is a broad range of phenotypes and exposures available for study to meet the goals of the RFA.

Would it be possible to combine smaller cohorts to meet the size requirements of the RFA?

Yes, applicants may propose a single study or a consortium of studies as needed to address the RFA goals of breadth and representativeness. Applicants proposing to combine multiple studies might consider including a management plan that explains to peer reviewers how funds would be distributed among the various other institutions and what the PI's role will be in leading this collaborative group. It will be important to justify to peer reviewers the merits of each of the particular cohort studies that were chosen to be included in the consortium.

Can we apply to this program with linked R01s or identify more than one Principal Investigator?

No, RFA HG-07-014 utilizes the U01 cooperative agreement mechanism and does not permit linked R01s or multiple Principal Investigators. Collaborators will need to select one Primary Investigator to be responsible for submitting the U01 application and managing and leading the collaborative study group if it is funded. PIs from collaborating institutions would be listed as key personnel on the grant application. This RFA is intended to support 3 to 5 study investigator groups (under RFA HG-07-014) and one coordinating center (under RFA HG-07-016). The PI of each of these study investigator awards along with the PI of the coordinating center and the NHGRI project scientist will serve as the voting members on the Steering Committee (SC). The SC will be responsible for coordinating the activities of the entire research program. Of course, the key personnel of each consortium will be able to attend the SC meetings, and participate in subcommittees, but only the PI will be a voting member.

Does the U01 mechanism have different requirements than the R01 mechanism?

Yes, the U01 cooperative agreement mechanism requires collaboration across funded sites and substantial involvement of NIH program staff. Specific requirements of the U01 are described in the RFA in section 2.A.3. (Collaborative Responsibilities), including a request for an indication of willingness of the applicant to be part of such collaboration.

What are the data sharing requirements for this research program?

Although NIH and NHGRI policies strongly encourage wide-spread sharing of de-identified genotype-phenotype data on individual participants through databases such as the controlled access portion of dbGaP, the sharing of individual-level data is not required for participation in this particular research program. Rapid and widespread sharing of the resulting summary genotype-phenotype descriptive and association data is expected, and is of high programmatic priority in selecting studies for award. Applicants should describe any potential barriers to sharing this summary level data as well as the barriers to sharing the individual-level data. See: RFA Section IV [Application and Submission Information (Part 6. Other submission requirements)] for a detailed description of what should be included in the application to address these particular data sharing issues.

Last updated: March 07, 2012