Under FNIH, GAIN will bring together the needed
scientific and financial resources through a
public-private partnership in which all potential
partners— including academia, government, private
foundations, advocacy groups, and others —will work
together toward a common goal. Pfizer Inc. is the
founding lead funder, having provided substantial
initial support for the development and implementation
of GAIN. Significant support has also been provided by
Affymetrix, Perlegen Sciences, The Broad Institute and
Abbott Laboratories. FNIH will continue to actively seek
partners for GAIN in order to extend support for
genotyping of additional diseases.
The
applications process for the first round of GAIN closed on May 9, 2006. There may be future rounds of
solicitation for the project, but they depend on obtaining additional financial support from potential
funding sources. Currently FNIH is accepting
Statement of
Interest which will help GAIN leadership plan for subsequent
application rounds should funds become available.
Top of Page
C. Selection of Projects
C1.
What is the scope of diseases considered eligible for
this project? What is meant by “common diseases?”
GAIN
does not have a specific set of diseases selected in advance, but will make decisions on particular studies
to be selected after considering the recommendations of the peer review committee and the Steering and Executive
Committees. Common diseases are considered to be diseases that have a significant impact on the public
health by affecting a significant percentage of the population.
Top of Page
C2. Who will determine which projects are genotyped?
The final decision rests with the FNIH Board. The decision
is based on input from the GAIN Steering Committee, the Technical Advisory Group, and the Peer Review Committee.
Top of Page
C3. What happens if my project is not selected for genotyping? Are there other funding opportunities for
which I might be eligible?
The FNIH is seeking additional financial support from a variety of potential funding sources in both the private
and the philanthropic sectors. If successful, FNIH will open additional rounds of solicitation for projects to
be genotyped under GAIN. And, of course, funding may be available from other sources, such as NIH.
Top of Page
D. DNA Submission and Genotyping for Selected Project
D1. Why is so much DNA needed? Can smaller amounts be accepted?
The genotyping will be carried out on high density arrays that will type at least 375,000 SNPs on each sample. To
be successful, the process will require a quantity of DNA that gives a high call rate (~99%). Inadequate
amounts of DNA will lead to the loss of too much data and may cause the entire array to give unusable data for
that sample. It is better to err on the side of too much DNA than too little.
Top of Page
D2. What do you mean by “high quality” DNA? How will quality
of DNA be assessed by the genotyping facility?
In
general, DNA to be genotyped must be of sufficient
concentration and sufficient size to give good results
with the genotyping platform. The actual techniques to
measure DNA concentration and integrity may vary
depending on which genotyping provider is used. Some
providers run agarose gels to assess the size (> 10
kilobases) of the DNA to ensure its quality. Other
providers may genotype a small number of STRP or SNP
markers on the samples to ensure that the quantity and
the quality of the DNA in each sample gives good call
rates. This latter procedure has the ability to identify
sample mix-ups if the study design is appropriate, e.g.,
the use of trios. Top of Page
D3. Who will pay for extraction of DNA and shipment of
samples to the genotyping facility?
The
investigator submitting the application will be
responsible for any costs involved in extraction of DNA
and shipment of the samples to the genotyping facility.
Top of Page
D4.
What genotyping quality procedures will be applied to
the genotyping? Will quality control information be
provided in the GAIN Database for each project?
The
genotyping quality procedures developed by the
Genotyping Working Group are available at
Genotyping and Analysis. Top of Page
D5. What will happen to any leftover DNA at the
genotyping facility?
The
genotyping facilities will ask only for an amount that
should give suitable genotyping results for the platform
to be used. It is expected that there will be little
excess DNA and that applicants will not want it to be
returned, but arrangements may be possible for return of
unused DNA if the applicant strongly desires it. Any DNA
not used for GAIN genotyping or returned to
investigators will be destroyed.
Top of Page
D6. Who will perform the genotyping and what methods will be used?
In this first round of GAIN Projects, genotyping will be
performed by Perlegen Sciences under agreement with
Pfizer and by the Broad Institute using genotyping
platforms donated by Affymetrix. 18,000 samples will be
genotyped. More information on methods may be found at
Genotyping and Analysis.
Top of Page
D7. Will the genotyping facility have access to any phenotypic data?
The genotyping facility will not have personal identifying information or advance access to any phenotype data or
personal identifying information. The genotype data and coded phenotype data will be merged in the GAIN Database
and will be made available through a Controlled Access process. Although genotype data on
individual study participants (without any link to individual phenotype or environmental data) will be
available on the GAIN Database within
dbGaP, phenotype data linked to genotypes will be available
only through the Controlled Access process. Investigators from the genotyping facility will have the
same rights to request access to these data as anyone else.
Top of Page
E.
Access to and Use of GAIN Data
E1.
Who will have access to data generated by GAIN, and how
will they access the data?
Both genotype data, and genotype data linked to coded phenotype data, along with pre-computed, simple
genotype/phenotype associations, LD measures, and information on additional genomic features will be made
available, through the GAIN Database within
dbGaP, to investigators whose requests for access have been
approved by the GAIN Data Access Committee (DAC). Applicants approved by the DAC will be given
password-protected access to the specific datasets within dbGaP for which the Data Access Committee has approved them. A
secure account with password and ID will be issued to the applicant by the DAC granting them appropriate
access to the GAIN Project Dataset within dbGaP.
If, and only if, existing consent within GAIN-selected studies is deemed adequate by the responsible
institution for open public access to genetic information, the genotype data generated for each sample
and summary information, such as allele and genotype frequencies, may be deposited in the public section of
the GAIN Database within dbGaP, with no individual identifying information and no connection to individual
phenotype data. Descriptive information of the phenotype data, such as the variables measured and protocols used
to collect the data and samples, but not the phenotype dataset itself, would also be made available in this
way.
In addition to the request and approval pathway for obtaining access to GAIN Project Datasets within the
Controlled Access portion of the GAIN Database within dbGaP, datasets may also be subject to Freedom of
Information requests. Should such requests be received, every effort will be made to identify alternative
mechanisms for providing access and to providing the government record information in a manner that is
consistent with ensuring participant privacy and data security.
Top of Page
E2. How quickly will genotype data from a selected project be posted to the GAIN Database?
The data will be made available as soon as the genotyping is completed and the data are cleaned and entered. This is
anticipated to be within 7 months from the time genotyping is initiated.
Top of Page
E3. What steps will be taken to limit access to the GAIN database and to protect it from potential misuse?
As noted in the answer to question E1 above, access to the GAIN database will be limited to specific projects
requested by the potential data user and approved by the Data Access Committee. In addition, both the requesting
investigator and a responsible Institutional Business Official from the investigator’s institution or business
entity must sign a Data Use Certification. Among other provisions, the Data Use Certification requires that these signatories
certify their acknowledgement of the GAIN
Policies, and their responsibility to comply with federal, state, and local laws and any institutional
policies; that the data will not be used to identify or contact individual participants from whom phenotype data
and DNA samples were collected; and that GAIN Project Datasets will not be transferred to anyone outside the
Requesting Investigator’s immediate research staff.
Finally, a GAIN Data Use Review Board (DURB) will provide ongoing, independent oversight of the data use and
handling policies and any participant protection concerns for GAIN project data in order to ensure that
all uses of GAIN-generated data conform to the specified research uses for each GAIN Project and adhere to the
highest possible standards for responsible research use, including the preservation of the privacy and
confidentiality of participants in the original parent studies.
Top of Page
E4. Will it be possible to withdraw from the GAIN Database any data from participants who have subsequently
withdrawn their consent for genotyping or data distribution?
It will be possible to remove data from the GAIN Database within
dbGaP if the Contributing Study Investigator
provides the unique sample code linked with the individual requesting withdrawal. However, once a
dataset in dbGaP has been accessed by researchers (using the appropriate access methods described in Question E1
above) it will not be possible to fully withdraw data from further GAIN-related research use.
Top of Page
E5. Are there privacy concerns related to posting an individual’s genome-wide genotype data on GAIN Database
within dbGaP, since it is a publicly accessible database?
Genotype data will be posted to the GAIN Database within dbGaP if, and only if, existing consent within
GAIN-selected studies is deemed adequate by the responsible institution for open public access to
genetic information. The individual genotype data that meet this strict criterion will be posted to the GAIN
Database within dbGaP without links to any identifying phenotypic information. Therefore, identifying an
individual in these limited cases based on GAIN genotype data would require matching those genotypes to another
genotype database that retains a link to personal identifiers--and such a linkage is very unlikely.
Top of Page
E6. What steps will be taken to ensure the security of the data associated with the project?
All data provided to FNIH in the application process will be placed on a secure server with limited access only to
FNIH and NCBI staff involved in the evaluation. Data on this server will only be backed up to a private tape
store for this project and will not be part of the long-term NCBI back up tapes or distributed outside the
secure area. All data will be destroyed after the application process except those for successful
applicants.
The genotype-phenotype data contained in the GAIN Database will also be maintained on a secure server with
appropriate firewall and security precautions, with access to the Controlled Access portion of the database
provided through password protected mechanisms as described in G1 and G3 above.
Top of Page
E7. Will there be public access to DNA extracted from GAIN samples?
There is no plan or provision for GAIN to distribute DNA samples. Contributing Study Investigators will send DNA
samples directly to the GAIN genotyping facilities. All DNA remaining at the conclusion of GAIN genotyping
activities will be returned directly to Contributing Study Investigators if requested, or destroyed as
described in F6 above.
Top of Page
E8. How will the databases used by GAIN actually be developed and maintained?
The core infrastructure for the central database will be developed and managed by the
National Center for Biotechnology Information (NCBI) on behalf of FNIH as part of the NIH contribution to
GAIN. NCBI activities will include the importing the coded phenotype data, and “cleaning” and entering the
genotype and pre-computed association data generated by GAIN into the GAIN database.
Top of Page
E9. Will any pre-analysis be performed on the data?
NCBI will perform a series of computations to determine genotype frequencies, Hardy-Weinberg deviations,
percentage of completed genotypes, and other indicators of data quality for all genotyping datasets. This
information will be posted on the GAIN Database within dbGaP once available. NCBI will also calculate
association statistics between genotypes and phenotypes as an element within the GAIN Project Datasets
available through Controlled Access portions of the GAIN database, as well as provide LD measures and additional
information on genomic regions associated with phenotypes. An Analysis Working Group will be formed by
GAIN to suggest analyses to be conducted across multiple projects and common approaches to analyses within each
project as needed to enhance comparability. NCBI may participate in this Analysis Working Group as
appropriate.
Top of Page
E10. Will the investigators submitting samples and data (the Contributing Study Investigators) be given any special
access to the database? Will they have any special rights related to the use of the data?
GAIN aims to release data to the scientific community as broadly and rapidly as possible (in a manner consistent
with its commitment to preserve the confidentiality of parent study participants) to maximize the public and
scientific benefits of the initiative. Therefore GAIN Project Datasets will be made immediately available for
access by users approved by the Data Access Committee. GAIN data will be made available for research to outside
investigators, GAIN Contributing Study Investigators, and GAIN Partners all at the same time and through the
same access approval mechanisms.
It is also important to recognize the important and unique contributions made by the Contributing Study
Investigators who have collected over many years the biological samples and associated phenotype data used in
GAIN. Therefore all outside investigators who apply to access the GAIN database will agree as part of the Data
Use Certification not to submit a publication on the data for a 9-month period from when the data are first
made available. Contributing Study Investigators will be exempt from this requirement, enabling them and their
direct collaborators a significant window for developing and submitting publications using the data from the GAIN
database. In addition, Contributing Study Investigators will have the benefit of participating in the Analysis
Working Group and Analysis Workshops.
Top of Page
E11. If information resulting from GAIN genotyping analyses suggests an association, but further research must be
conducted for validation, how would those costs be covered?
GAIN will not cover these costs; additional funding should be sought through the internal funding mechanisms of the
investigator’s organization or through the usual funding mechanisms at NIH, other government agencies,
professional organizations, and/or non-profit foundations.
Top of Page
F. Human Subjects
F1. Has the Office for Human Research Protection
(OHRP) reviewed the GAIN data submission and release
policies and procedures?
The Office for Human Research Protection (OHRP) reviewed the GAIN data submission and release policies and procedures
and concluded that the GAIN Database does not constitute human subjects research, since GAIN will not receive the
key-code to identify any individuals within the GAIN Project Dataset(s) from GAIN Contributing
Investigators. Furthermore, investigators requesting access to GAIN Project Datasets will only receive
de-identified data from the GAIN Database, and therefore will not be conducting human subjects research as
defined within 45 CFR Part 46. This determination does not negate any existing local (institutional or
state-level) requirements for the on-going review of research. (OHRP Letter to GAIN)
Top of Page
