You are here

Instructions for ADCs

Instructions for Access to Sequenced Subjects from your Alzheimer’s Disease Center (ADC) through NIAGADS

Last Updated 10/02/2018

Background

This document provides information on how to request access to Alzheimer’s Disease Sequencing Project (ADSP) data for subjects from your Alzheimer’s Disease Center (ADC) being distributed through NIAGADS DSS.

Many samples included in the ADSP whole genome sequencing (WGS) and whole exome sequencing (WES) study were contributed by the National Institute on Aging (NIA) funded ADCs. The ADSP has released WGS data for 3,941 subjects and WES data for 10,914 subjects to the research community through dbGaP (hg37 data) and NIAGADS DSS (hg38 data). Please see the ADSP website for more details on the study.

When accessing the data, every effort should be made to comply with the NIH Genomics Data Sharing Policy http://gds.nih.gov/, taking special note of the following provision in the policy:  

4.  Non-Identification

Approved Users agree not to use the requested datasets, either alone or in concert with any other information, to identify or contact individual participants from whom data and/or DNA samples were collected. This provision does not apply to research investigators operating with specific IRB approval, pursuant to 45 CFR 46, to contact individuals within datasets or to obtain and use identifying information under an IRB approved research protocol. All investigators conducting human subjects research² within the scope of 45 CFR 46 must comply with the requirements contained therein.

If your research does not require matching of ADC subjects, please follow the Instructions for Application (https://dss.niagads.org/documentation/applying-for-data/application-inst...) to submit your NIAGADS DSS Data Access Request.

Specific Information for Requesting Data

  1. To access ADSP data all investigators must complete a Data Access Agreement with the University of Pennsylvania. https://dss.niagads.org/wp-content/uploads/2018/10/NIAGADS_DAA_ADCreturn...
  2. To ensure proper de-identification of individual subjects in the study, each ADSP subject is assigned a unique ADSP ID.
  3. In order for ADCs to link ADSP data to their own Center’s clinical records, the ADSP data need to be matched. This is done using the original patient IDs (PTIDs) from the contributing ADC.
  4. IRB approval is needed in order to match the ADSP IDs to the subjects from the Investigator’s Center (and only from their Center).

Overview of the Required Steps

The rest of this document outlines steps required to access and match ADC subject data with ADSP sequence data.

All subjects in the ADSP dataset are assigned a unique ADSP subject ID that can be used for matching. The information for mapping ADSP IDs to the original patient IDs from the contributing ADCs is available from The National Alzheimer’s Coordination Center (NACC). Go to https://www.alz.washington.edu/GWASPHASE2/adsp.html to obtain this key. For each ADC, NACC will provide ID mapping information ONLY for subjects from that specific ADC.

Each ADC should follow the 3 steps below to access and match ADSP data to its subjects.

  1. Submit research protocol to your institutional IRB for approval and send to NIAGADS for review.
  2. Request data from NIAGADS and complete a Data Access Agreement with the University of Pennsylvania.
  3. Obtain ADSP ADC Patient ID mapping information from NACC.

 

Step 1. Prepare research protocol for IRB approval and send to NIAGADS

ADSP sequence data are de-identified when they are released. Subjects cannot be matched without having explicit knowledge of the subject’s identity, such as mapping information for the original subjects or extensive genetic knowledge from the subject or his/her relatives.

If you plan to ascertain data for subjects whose DNA has been sequenced by the ADSP, you will need a full IRB review and explicit permission from the IRB to do so. If you only plan to work on de-identified data, an expedited review may be sufficient. The IRB from your institute has the authority to approve or deny your proposal. Please send a copy of your IRB approval to NIAGADS for review.

What information should be included in your IRB protocol? A typical IRB protocol should describe the research plan, subjects being studied, and a benefit/risk analysis as justification of the research plan.

1. Which subjects sequenced by the ADSP have data that will be matched in this protocol?

If you follow the matching procedure we described in this document, you will only match subjects from your own ADC.

2. How will the subjects be matched?

Your IRB protocol should describe the workflow to match relevant subjects in the ADSP data. The following paragraph describes the procedure planned by NIAGADS, NCRAD, and NACC and outlined in Steps 1, 4, and 5. If your workflow is different, you should edit or prepare your own description in your IRB protocol.

We will receive the key to map ADSP IDs to the original patient identification IDs (PTIDs) from the National Alzheimer’s Coordination Center (NACC). We will receive the mapping key only for subjects recruited from our own ADC. The key has been prepared by NACC and the NIA Genetics of Alzheimer's Disease Data Storage Site (NIAGADS). We will receive ADSP individual level data from the NIAGADS at the University of Pennsylvania, then use the mapping key to match subjects to our PTIDs in the sequencing data and genotype data. Only subjects from our ADC will be matched.3. Why is matching necessary? What is the risk for the subjects who are being matched? (Benefit versus Risk of the research project)

The purpose of matching the ADSP subjects from your ADC is to link the ADSP data with clinical records on your subject. If your research can be carried out without identification of the ADSP subjects, you should avoid matching the subjects altogether. If matching is necessary, you should provide a good scientific justification for proposed research activities.

What about incidental findings? Incidental findings are genetic findings that are irrelevant to the purpose and scope of the proposed analysis. Sometimes these incidental findings may have clinical implications (e.g. genetic variants that have high predisposition to diseases such as certain BRCA1 mutations that may cause breast cancer) and/or ethical ramifications (e.g. previously unknown relatedness or non-relatedness to other individuals). The ADSP Memorandum of Understanding (MOU) (https://www.niagads.org/adsp/content/about) states that incidental findings will not be revealed to the subjects.

 

Step 2. Request data from NIAGADS and complete a Data Access Agreement with the University of Pennsylvania.

Send an email to niagads@pennmedicine.upenn.edu to initiate a request for data. This email should include the Alzheimer’s Disease Center Director from the requesting Institution. Identify the ADC and supply a signed copy of the Data Access Agreement (DAA) to be countersigned by the University of Pennsylvania. A copy of the DAA template can be found here: https://dss.niagads.org/NIAGADS-DAA-ADCreturn.pdf.

Work with NIAGADS to identify the data types you would like to receive and options for data transfer. All sequencing data will be mapped to GRCh38. Available data types include CRAMs, gVCFs, and joint-called genotypes in a project level VCF.

 

Step 3. Obtain ADSP ID to ADC Patient ID mapping from NACC

NACC will provide each ADC with the ID mapping information for the subjects contributed by that specific ADC.

Mapping Information can be downloaded from https://www.alz.washington.edu/GWASPHASE2/adsp.html using your center log-in information. Please contact NACC (naccmail@uw.edu) if you have questions or if you do not have a NACC web account.

Each sample contains the following information:

  1. The ADSP Subject ID for the subject being sequenced.
  2. The ADSP Sample ID for the DNA sample being sequenced. The Sample ID contains the corresponding Subject ID as part of the full ID text.
  3. The original Patient ID (PTID) assigned by the contributing ADC.

Theme by Danetsoft and Danang Probo Sayekti inspired by Maksimer