Data

Analysis Proposals

 

  1. Purpose:  To promote collaboration and wide use of Alzheimer Disease Genetics Consortium (ADGC) data and to further Alzheimer’s disease research, members of the ADGC can submit proposals to conduct analyses using data from multiple cohorts which have been assembled by the ADGC for genetic studies.  These proposals are called Special Analysis Group (SAG) projects and are considered collaborative efforts.

 

  1. Eligible Organizations/Principal Investigators (PI):  The PI of the proposal must be a member of an organization that participates in the ADGC such as a NIA-funded Alzheimer Disease Center.  The proposal must include a member of the ADGC Analysis Committee (ADGC-AC, co-chaired by Drs. Pericak-Vance and Farrer) as a consultant (list available on the ADGC web site (link).   Alternately, if the PI of the proposal is not currently a member of the ADGC or at an organization that participates in the ADGC, a proposal can be submitted as a collaboration with one of the PIs of the ADGC (Gerard D. Schellenberg, Lindsay Farrer, Jonathan Haines, Richard Mayeux, Margaret Pericak-Vance, and Li-San Wang, or their designee).  The ADGC consultant or ADGC PI associated with a SAG will monitor progress and ensure that the project conforms to the original description submitted in the SAG proposal.  The ADGC consultant/PI may or may not be co-authors on the final published studies, depending on their contribution.  For projects where the ADGC supplies data, the “ADGC” will be listed as a co-author on the title page.

 

  1. Timeline and review: Proposals can be submitted at any time.  The analysis proposals will be assigned to two members of the ADGC-AC, and the review will be completed in approximately 2 weeks.  The proposal will then be discussed by the entire ADGC-AC.  The Committee meets on the second and fourth Wednesday of each month.  The original reviewers or the ADGC-AC may request additional information or changes as needed and then re-review the proposal.

 

If no changes are needed, the proposal will be circulated to the ADGC Executive Committee with a 48-hour time period for comments.  If no objections are received, the SAG proposal will be considered approved.  If minor changes are recommended by the ADGC-AC, the SAG PI will make changes in consultation with the ADGC PI or the ADGC-AC participant, and changes returned to the original reviewers.  If the changes are accepted by these reviewers, the proposal will then be circulated to the ADGC Executive Committee for a 48 hr. comment period.  If no objections are received, the SAG is approved.  If major changes are required, the SAG will be re-reviewed by the two original reviewers, and then discussed again by the full ADGC-AC. 

 

  1. Restrictions: Applications should not overlap with core analyses that are being performed by the ADGC Analysis groups. The ADGC core analysis projects are listed on the ADGC web site (http://www.adgenetics.org/content/adgc-core-projects).  If a SAG overlaps with another previously approved SAG proposal, the investigator will be encouraged to contact the PI of the conflicting proposal with the goal of working together. A list of approved SAGs can be viewed on the ADGC web site (http://www.adgenetics.org/content/current-adgc-special-analysis-group-pr...)

 

  1. Data sets for SAG analyses:  The ADGC will provide the most recent and most comprehensive data available. The data sets available are listed on the ADGC Web site (link).  These data will have been QC’ed by members of the ADGC-AC using a uniform process.   As noted in the list of cohorts posted on the ADGC web site, some datasets require permission of the PI who contributed the dataset, and it is the responsibility of the SAG PI to seek permission from those PIs to use their data.   If requested, imputed data will be provided using the most recent and largest imputation panel (e.g. TOPMed).  A minimum phenotype dataset will be provided.  If additional phenotypes are needed, the ADGC will work with the investigator to identify and acquire the needed data. 
  2. Application:  Investigators wishing to submit a SAG proposal are encourage to contact Marissa Cranney (marissa.cranney@pennmedicine.upenn.edu) for guidance on the submission process.  The final SAG should be submitted to Marissa Cranney to start the review process.
  1. Prior to submission:
  1. Contact Marissa Cranney to gain access to needed material (get access to ADGC member web site).
  2. Review core ADGC analyses to avoid overlap.
  3. Review approved SAGs for possible overlap or collaborations.

 

  1. Format: Proposals should be up to 4 pages and should include the following:
  1. Analyses to be performed (1 ½ pages)
  2. Datasets to be analyzed and format, i.e. imputation (see note below regarding data access)
  3. Variable(s) to be requested and analyzed
  4. Members of the proposed analysis teams
  5. Data to be added (e.g. expression data, eSNP data, data from other studies, etc.)
  6. Data cleaning methods.  The ADGC has a standard QC pipeline   If the applicant wants to receive unprocessed pre-cleaning data, please justify this request in the proposal
  7. Population sub-structure analysis (if receiving unprocessed data)
  8. Timeline
  9. Deliverables
  10. NIH Bio sketch for the PI and key collaborators
  1. Guidelines for analysis of ADGC data:

 

  1. Data provided by the ADGC can only be used for the proposed analyses.  If the goals and aims of the project change, the SAG proposal must be amended and submitted for review.

 

  1. PIs of approved analyses will communicate with the ADGC consultant/PI quarterly? to give periodic updates on progress.
  2. PI will give a progress report on an ADGC-AC call prior to drafting a manuscript. Contact Marissa Cranney to schedule.
  3. PI will provide a draft manuscript to the ADGC for review, (at least 2 weeks?) prior to submission to an online or print journal. The results will be published acknowledging the ADGC as a co-author and listing appropriate grants.
  4. All genotype/phenotype data must be kept confidential and can be distributed only to consortium members who are part of the proposed project analysis team (no secondary sharing unless specified in the SAG proposal).
  5. Multiple groups may propose the same analyses.  If this occurs, all groups will present their results and be encouraged to prepare a joint publication. 
  6. Summary statistics should be submitted to the NIA Genetics of Alzheimer’s Disease Data Storage Site (NIAGADS).
  1. Final Comments.  SAGs are a collaboration between the SAG PI and the ADGC.  The goal is to get the most recent and complete harmonized data to investigators for genetic analyses relevant to ADRD.  The ADGC attempts to foster collaborative team science while minimizing barriers to data access.  Open communication is encouraged as is collaboration between groups with common interests.  The ADGC is a community of collaborative investigators operating under an “open science” model.