Join me in voicing support for including data sharing plans and track record as criteria in NIH grant review. Currently the NIH requires a data sharing plan for large grants, but explicitly excludes reviewing this plan as part of assessing the scientific merit of a proposal. This should change.
The email below was sent to the iscb-publicaffairs-updates list. Thanks to Dr David States a) for championing these causes and b) drafting language for us to reuse in voicing our support.
Email your comments to PeerReviewRFI@mail.nih.gov by March 17, 2008 (or better yet, do it right now!)
While you’re at it, the ideas below about web resources sound great too.
The US NIH is in the final phase of revising the peer review process by
which grants are reviewed and awarded. See
http://enhancing-peer-review.nih.gov/ for details and a link to the
report. The Final Draft Report identifies “the most significant
challenges facing the NIH peer review system” and proposes recommended actions.
As the new chair of the ISCB Public Affairs & Policies Committee, there
are two specific issues that I would like to bring to your attention in
the proposed revisions to the NIH peer review process that are of direct impact to computational biologists:
1) Access to URLs and web materials in the review process. The current NIH guidelines discourage applicants from including URLs in proposals and discourage reviewers from accessing web sites in the review process.
The basis for this policy is two fold, most importantly protecting the
reviewers’ confidentiality but also that a proposal needs to stand on
its own. In my view the latter consideration needs to be tempered by
the fact that reviewers look at publications from an applicant all the
time. Looking at a web site is no different from looking at a paper to
clarify a concern. This policy can work to the disadvantage of
bioinformatics projects where the project web site is an important
mechanism for dissemination and data sharing.
Proposal: NIH should establish an anonymous web proxy server for use by reviewers so that they can visit and evaluate web services and web content described in a proposal. Applicants should be encouraged to include URLs as evidence of project performance.
2) Data sharing. The sharing of data by an investigator should be
included as a review criteria and there should be a section of the
proposal addressing data sharing plans. Data sharing is not an all or
nothing issue. There are issues about how the data is made available,
documented and in the case of web servers what restrictions are placed
on access and downloads. I think it is appropriate for the reviewers to
consider data sharing behavior in evaluating the merits of a proposal.
Proposal: Data sharing behavior should be made an explicit review
criterion. There should be a section in all proposals discussing data
sharing track record and plans. For renewal applications, there should
be a section in the proposal with actual accessions and URLs for data
that has been deposited with repositories or made accessible on the web.
These are my personal opinions and not necessarily those of ISCB, but if you share my concerns and wish to use any of my text above in your own comments, please feel free to do so.
David J. States, M.D., Ph.D.
Chair, ISCB Public Affairs & Policies Committee