TOPMed Publications Policy

To facilitate the use of TOPMed data to support scientific discovery while protecting TOPMed studies and participants, and to minimize burden on applicants and reviewers, the TOPMed Publications Advisory Committee and Executive Committee (EC) have established the following Publications Policy. The policy is intended to ensure transparency, enable productivity tracking, and promote synergy across the TOPMed projects and studies.

This Policy applies to any use of TOPMed data by a TOPMed investigator, including phenotypic, genotypic, or other molecular data. This includes pre-release data obtained through the Exchange Area using the TOPMed-generic Data Access Request process, as well as to pre-release data obtained through other means, e.g., directly from a sequencing/omics center or from the TOPMed Informatics Research Center (IRC). The Policy also applies to TOPMed investigators’ use of released TOPMed data obtained through a standard (vs TOPMed-generic) dbGaP application.

This Policy does not apply to the following activities:

1. Proposals, manuscripts, and abstracts from non-TOPMed investigators, which necessarily use only released TOPMed data obtained through a standard (vs TOPMed-generic) dbGaP application.
2. Proposals, manuscripts, and abstracts from TOPMed investigators where the only use of TOPMed data is variant summary information from the BRAVO (Browse All Variants Online) server.  
3. Proposals, manuscripts, and abstracts from TOPMed members where the only use of TOPMed data is imputation to the TOPMed reference panel constructed by the IRC and made available on the Michigan Imputation Server. Notably, such products should acknowledge the TOPMed panel, including the TOPMed Program, contributing studies, the IRC, and the Administrative Coordinating Center (ACC). 
   • However, note that uses of a locally constructed (vs. the IRC’s) TOPMed reference panel using pre-release data are within the scope of this Policy.
4. Proposals, manuscripts, and abstracts from either TOPMed investigators or non-TOPMed investigators that use Genomic Summary Results (GSR) from a published TOPMed manuscript.
   • However, note that uses of unpublished TOPMed GSR are within the scope of this Policy.
5. Proposals, manuscripts, and abstracts led by non-TOPMed investigators for which one or more TOPMed investigators provide consultation and input that does not use data or leverage their position in TOPMed.
   • However, if a TOPMed investigator’s contributions to an external paper substantially leverage the investigator’s role in TOPMed, e.g. through specific Working Group knowledge or data access, then the activity is within scope. A TOPMed investigator will need to assume responsibility for carrying out TOPMed publications procedures for the paper unless there are additional reasons to onboard the external investigator such that they are able to carry out those processes directly.
   • Should there be uncertainty about whether the policy applies, the investigator is encouraged to contact the Publications Advisory Committee.
6. Proposals, manuscripts, and abstracts where the only use of TOPMed data is for infrastructure development and/or testing in the BioData Catalyst cloud ecosystem.
   • TOPMed investigators involved in BioData Catalyst infrastructure development and/or testing should refer to BioData Catalyst Consortium publication policies and tracking procedures.
   • In contrast, analytical or hypothesis-generating scientific work by BioData Catalyst developers is within scope of the TOPMed Publications Policy if the work uses TOPMed data and the developers are also TOPMed investigators.

Notably, it is incumbent upon TOPMed investigators to be aware of and follow additional policies that apply to their activities, whether within or beyond the scope of this Publications Policy. Such additional policies include the National Institutes of Health Genomic Data Sharing Policy, dbGaP policies, the Genomic Data User Code of Conduct , and any other local or institution-specific policies.

Priority will be given to manuscript proposals that address the original aims of the studies that were selected for inclusion in the TOPMed program. Other within-study or cross-study proposals will also be reviewed. Cross-study proposals that maximize the value of the TOPMed program as a whole are strongly encouraged.

Working Groups (WG) should involve all investigators, particularly junior members, as early as possible in analyses, and provide opportunities for them to take active roles, including leadership, that merit authorship.

Manuscript proposals are typically developed in the context of one WG (i.e., the primary working group). Studies whose primary approach spans multiple WGs choose a primary WG, but should also engage the other relevant WGs. Proposals generally should comprise a focused pre-defined project (e.g., Whole Genome Sequence Analysis (WGSA) for discovery of variants associated with phenotype X). The primary working group assists in the scientific development of project proposals in collaboration with the leads for the project, and when appropriate, sequencing and prioritization of all working group projects to build an overall arc of discovery for that area. Working group conveners are encouraged to engage as many cohorts, analysts, and project leads as possible to maximize productivity and opportunity for all contributors. Proposals should be focused and provide opportunities for various (usually junior) investigators to take a lead. However, working group convenors should weigh carefully the ability of a given (lead) junior investigator to manage more than two active projects. Each proposal typically will focus on one phenotype, or one specific hypothesis / analysis, and result in one manuscript. Omnibus proposals that encumber many phenotypes and many analyses or those that are anticipated to result in several manuscripts are discouraged, and proposers will be asked to break down proposals into components.

Data use requests are made to complete the proposal preparation. The PI of each participating study will be petitioned for permission for data use. As part of this process, authorship plans should be discussed (see below). 

Manuscript proposals are submitted to the Publications Committee by following these instructions for paper proposal development and submission. Approval of the primary working group is needed at the time of submission of the proposal to the TOPMed Publications Committee (P&P). Working group convenors are expected to respond to proposers within two weeks, and schedule a discussion of submitted proposals. Proposers may report delays to the ACC (topmed-pubs@westat.com), who will investigate and, if necessary, consult with the P&P and/or EC.  

With P&P approval and data use requests granted, data analysis and interpretation commences. If the project evolves in approach, lead author, or contributors over the course of development, the project lead should amend the proposal to reflect those changes; the final proposal will be used in administrative checks prior to journal submission of the resulting manuscript. If substantive changes to scope or focus occur, a major proposal revision should be submitted for re-review.

For the duration of the TOPMed program, any submitted abstract or manuscript using TOPMed data must be associated with an approved manuscript proposal. All manuscript proposals must be reviewed by an appropriate WG and submitted for P&P. The P&P administrator will be responsible for logging each project proposal, and assigning a number (via the web-based proposal submission process). The proposal will then be available to all P&P members for review and comment for a period of two weeks. At the end of the comment period, the applicant will be informed of the P&P’s decision. Possible decisions include: approve, reconsider after revision, conflicted, or disapprove. Any disagreement between the P&P and an applying investigator(s) will be referred to the EC for resolution, whose decisions will be final.

The P&P will periodically review and request updates for approved manuscript proposals. It is the responsibility of the WG Conveners to monitor progress of ongoing approved proposals. Proposals with no progress for a six month period, particularly those that are causing delay of other projects led by junior investigators, may be subject to review and re-assignment of the manuscript, restructuring of the writing group, or other actions.

The following section applies to “single PI” proposals and manuscripts. For the purposes of the TOPMed publications policy, a “single PI” proposal or manuscript is defined as one that (a) uses data from one or more TOPMed parent studies led by a single TOPMed PI and (b) is proposed by or with the approval of that PI. Multi-PI or co-PI collaborations may also be treated as a “single PI” in certain cases, such as when the component studies are part of a single TOPMed parent study or X01 grant. 

The ACC (with input from the Publications Advisory Committee) can advise proposers unsure of whether they may fit this “single PI” class of proposals. 

In general, the policies and procedures described throughout this document apply to “single PI” proposals and manuscripts, with the following variations:

• Working Group approval is not required for “single PI” proposals or manuscripts in order to submit to the TOPMed system. However, proposers are required to circulate these products to the WG for informational and tracking purposes. Proposers are also encouraged to present to the WG at the discretion of the convener(s).
• Scientific review for “single PI” papers is advisory only. Specifically, single study proposals are not required to respond to requests for revision by the P&P committee, and the ACC will automatically approve their scientific review regardless of the comments.
• Dataset selection request is still required of “single PI” proposals, as a safeguard to make sure that dataset contacts are aware of each requested use.
• Manuscript review is limited to administrative checks performed by the ACC.

Including “single PI” proposals and manuscripts in these policies and procedures is important for tracking productivity, promoting transparency within the TOPMed program, and ensuring that study investigators are aware of and approve each use of data from their study.

Where possible, all studies that have the phenotypes needed for a proposed manuscript should be offered a chance to participate. No study’s data may be used in a manuscript without an offer of collaboration and the study PI’s approval for data access. Investigators should involve in the writing group authors from studies whose data are included in a manuscript as well as members from the ACC, IRC and sequencing/omics centers – particularly those who take responsibility for ensuring manuscripts describe data generation and cleaning accurately, and with relevance to the analysis at hand. Manuscript proposers are expected to initiate contact with study PIs as well as the ACC, IRC, and sequencing/omics centers to negotiate a fair number of authorship slots and request appropriate co-author representation. The specific authors can be nominated with the advice of the study PI initially, and may be revised based on the contributions of the investigators to the advancement of the project. Final authorship list should be verified with the study PI at the manuscript stage. TOPMed provides guidelines for authorship including use of banner authorship.  

A study may decline to collaborate in any manuscript. However, given the substantial investment that has been made in TOPMed sequencing, studies should justify the reasons for withholding data, and, if applicable, appropriate conditions or time frame when such data can be shared. Data should not be withheld from use in a manuscript beyond the six month embargo period, save for reasons such as those related to participant consent, potential harm to participants, or other substantive issues, subject to discussion.

Abstracts must be associated with at least one TOPMed proposal. Exceptions must be approved by the P&P Committee and communicated to the ACC prior to submission. A single TOPMed proposal may be associated with multiple abstracts as long as all of these abstracts result in one published paper.

Abstracts must be submitted to the P&P by using the Abstract Submission Form. Abstracts are not reviewed by the P&P.

Submitted abstracts can be viewed with the online Abstract Viewer.

Manuscripts will be reviewed and approved by the WG conveners prior to submission. Every effort should be made to confirm authorship prior to submission. The ACC will provide an administrative review of the manuscript, based on the original or amended proposal. Please review the details on the Manuscript Writing and Submission Instructions for more information.

Studies whose policies require study-specific review by their own publications committees are requested to complete review of abstracts and manuscripts within two weeks, with a focus on acceptable authorship, affiliations, study description, and funding acknowledgements. Upon submission of the manuscript to the TOPMed website, lead authors will be required to provide either study-specific approvals or documentation of their communication efforts to solicit approvals at least two weeks prior.

As stated above, manuscripts must be associated with a previously approved TOPMed manuscript proposal. Closely timed submission of a manuscript proposal and the related abstract or manuscript will be viewed as lacking transparency, is strongly discouraged, and may lead to delay in manuscript approval and submission.

Completed manuscripts should be submitted for administrative review at least ten business days before anticipated journal submission to ensure complete and accurate funding acknowledgements. Lead authors are requested to notify the ACC when a manuscript is accepted for publication.

Visit the TOPMed archive of manuscript proposals.

Please refer any questions about TOPMed publication policies and procedures to Publications Advisory Committee Chairs and the ACC (topmed-pubs@westat.com).