TOPMed Authorship Guidelines

1. All contributing studies should be asked whether they wish to contribute a co-author. Some studies require a co-author and, in that case, they must be included. Please review the list of studies that require a co-author.

2. Authors representing the IRC and sequencing centers for the data used in the analysis generally should be included.

   1. Please see the key contacts page for relevant authorship contacts.

   2. Please see the authorship guidelines for contributions to identify relevant citations and determine if you should include DCC authors.

3. The TOPMed banner, "NHLBI Trans-Omics for Precision Medicine (TOPMed) Consortium", is required in the authorship list, when allowed by the journal.  This requirement applies to all multi-study papers in which the primary use of TOPMed data is for discovery or for methods development. 

   1. Please include the full TOPMed Banner Authorship list as Supplemental Material, or as the journal directs. However, if the journal does not allow this special class of authors, they should be acknowledged with a link to the banner author list on the TOPMed web site in the acknowledgement section of the paper.

   2. Study-specific banners should not be listed in addition to the TOPMed banner. If requested, study-specific banners for the use of TOPMed data can be listed in the acknowledgements, not in the author list.

   3. The position of the banner text within the byline author list is at the discretion of the authors (i.e., it could be last or somewhere in the middle).

   4. You may also list a TOPMed Working Group (WG) in the byline, but this does not replace the requirement for the general TOPMed banner.

   5. Papers based largely on other sources of data, but in which TOPMed provides replication or other secondary analysis results, are not required to use the banner. TOPMed acknowledgements should be included.

   6. While not required, banner authorship should also be considered for single PI papers.

4. Authors should review and adhere to the authorship guidelines of the journal to which they are submitting. Specifically, many journals have signed onto the authorship criteria recommended by the ICMJE (International Committee of Medical Journal Editors). As noted above, for journals that do not allow banner authorship, the banner can be linked  in the acknowledgements section.

5. The above guidelines generally apply to TOPMed abstracts as well as manuscripts; however, it is expected that the number of authorship slots on abstracts may be limited (e.g., by conference organizers). In the case of limited authorship slots on abstracts, it is recommended to prioritize direct contributions and include the TOPMed banner to cover other contributors. As a courtesy, all co-authors on the paper proposal on which an abstract is based should be notified of the abstract submission.

The goal is fair scientific representation from TOPMed investigators and contributing members (if any) of external, non-TOPMed groups offering bespoke annotation or physiological, functional or replication data. As a general rule, early career (‘junior’) investigators will occupy first authorship positions.

We recommend that WGs consider the following guidelines for authorship for collaborative manuscripts:

Multiple first and multiple senior authors may be designated as such. Whenever possible, junior investigators will occupy first authorship positions. Senior authors will be listed in full at the back of the author list in print versions of manuscript(s), and all by-line authors will appear as Authors in PubMed. Senior author order from the last position is dictated by three major considerations: a) their direct intellectual participation in the drafting and completion of the manuscript, b) achieving balance with the junior authors positioned at the front end of the authorship list so as not to overemphasize a single group, and c) the number of genotypes/phenotypes contributed by any samples they represent.

Long lists of authors are permitted if all authors meet standards for authorship as defined by ICJME standards.

Authorship position will be determined in advance by a group consensus by the researchers actively participating in the paper describing a specific study. Some of the factors that may be considered include effort in the WG for analyses and writing, contributions by size of the contributing TOPMed study, and phenotype data quality and quantity. For any authors representing studies contributing in silico replication data, similar authorship positions apply. For any authors representing groups contributing bespoke physiological or functional data, their magnitude of contribution in these spheres will be considered, again with junior authors preferred for “front” positions.

The general order of authorship will be, from First to Senior:

1. Writing team members (expected to number about 4 to 10, with author order and any starred First authors to be chosen in draft form by these ~4-10 individuals)
2. Individuals who made significant analytical, or physiological or functional contributions from either discovery or replication stages, but not listed in #1. Generally in the following order:
   
   1. Individuals who performed analyses of the whole- genome sequence or other omics data-wide analyses, not in #1
   2. Individuals who performed additional annotation, physiological or functional analyses, not in #1
   3. Individuals from non-TOPMed consortia who performed replication look-ups, not in #1
3. Alphabetical listing for ‘middle authors’ not in #1-2 or #4-6
4. PIs from physiological, functional or replication consortia, projects, studies, or labs
   
   1. May include individuals contributing look-up data from other TOPMed phenotypic working groups (i.e. a diabetes paper with using hematology WG data and others).
5. Other PIs from TOPMed discovery studies, minus those listed in #6; senior representatives from IRC and sequencing / omics centers not included in #6
   
   1. May include individuals contributing look-up data from other TOPMed phenotypic WGs (i.e. a diabetes paper with using hematology WG data and others).
6. Senior authors representing the relevant TOPMed WG and its constituent TOPMed projects and studies

Authorship order for #1, #5, and #6 above will be initially tasked to the writing team members and will reflect the contributions of each author and group to the paper. Final order will be approved by the PIs representing TOPMed discovery studies and TOPMed WG leadership.

The number of authors per contributing TOPMed study or replication study should generally reflect the contribution of the study to the paper. A range of three to seven authors from any individual study is a reasonable guideline.

The number of authors allowed from other TOPMed phenotypic WGs or from other studies or consortia providing physiological data or ‘look –up’ replication data will be determined by negotiation with those groups. It is expected that at least the junior analysts doing additional experiments, analyses or look-ups, plus a limited number of senior investigators from those groups would be included.

In situations of authorship dispute, leadership of the individual TOPMed WG will moderate discussions, except for cases where the WG leaders have a conflict of interest in authorship. In these cases, involvement of TOPMed Publications and / or the Executive Committee will be considered.