The NIHghts who say 'no' – to chemoinformatics

A recent post from The Sceptical Chymist: The NIHghts who say ‘no [1]

The NIHghts who say ‘no’

Apologies to our international readers for the U.S.-centric post, but the National Institutes of Health announced earlier today that PAR-07-353, a grant involving Cheminformatics Research Centers, has been canceled for “programmatic reasons.” For those of you who haven’t heard of the Cheminformatics Research Centers, they are part of the Molecular Libraries Roadmap Program (MLP), which is

an integrated set of initiatives aimed at developing and using selective and potent chemical probes for basic research … [The MLP] was proposed to introduce high-throughput screening approaches to small molecule discovery, formerly limited to the pharmaceutical research industry, into the public sector… [and] is made up of the following major components: (1) access to a library of compounds (Molecular Libraries Small Molecule Repository); (2) access to bioassays provided by the larger research community; (3) support for the development of breakthrough instrumentation technologies; (4) access to a network of screening and chemical probe generation centers (MLPCN) where assays are screened and probe development is undertaken; (5) Pubchem, the primary portal through which the screening results of the MLPCN are made public and (6) the Cheminformatics Research Centers (CRCs) with multiple roles focused on high-level data analysis and dissemination with a focus on developing new understanding of the cellular processes (genes and pathways).

One reason why this is so surprising is because the grants were due next week (June 28th). I imagine the timing of this decision (and the decision itself) is bound to upset a number of people in this community, especially since many applicants were probably working around the clock to get their grant submitted before the (now non-existent) deadline…
Does anyone know more about this story or why the grant was canceled?
Joshua Finkelstein (Senior Editor, Nature)

First, Joshua, no apologies needed – this affects world science not just US chemoinformatics. (And a reminder that Nature was active in helping to report the activities of those of us who wished to promote the value of the Pubchem effort). In Cambridge we are (or were) working with potential applicants in this program and were in the process of preparing material. We were informed:
There is a notice published today in the NIH Guide
that cancels the “Preapplication for Cheminformatics Research Centers
(X)2) PAR-07-353. Here is the essential element of that notice:
“This Notice is to inform the scientific community of the cancellation
of the PAR-07-353 entitled, Preapplication for Cheminformatics Research
Centers (X02), due to programmatic reasons. Applications should not be
submitted for the June 28, 2007 date. Any application submitted will not
be assigned or reviewed. NIH intends to re-issue this announcement at a
later date.”
The point is that NIH is (or was until yesterday) reaching out to the world community. Chem(o)informatics is a key tool in understanding biology and in discovering new leads for pharma. There was a real chance to revitalise chemoinformatics which has been languishing for many years bedevilled by lack of:
  • Open Data
  • Open software
  • Open processes
  • a modern approach to information
  • reproducible science
Just last week we had a 3-day workshop at the Unilever Centre on Machine Learning, aimed primarily at chemoinformatics. One of the speakers, a statistician, has published concerns about the quality of science in many chemoinformatics publications – it suffers from all the above list.
The NIH program would have given a major boost to Open reproducible science. The data would have been fully Open and reusable, software would have been Open and modular, and chemistry would have had a major example of how data could be re-used for science rather than being aggregated and resold. Whether or not our group had been funded I was looking forward to this program as it would have been a highly cost-effective use of funds. And it could have shown the pharma industry, which relies heavily on this approach but does so little to encourage good practice in it, a way forward. Will any pharma CEOs speak out? Or do anything to help a science on which they depend?
It would be irresponsible to speculate on what “programmatic reasons” means. It could be that, like Britain, the US wants to spend its income on wars rather than health. But unfortunately not every US organization approves of the NIH’s funding of Pubchem and related projects which are often seen as “socialist” and seen as the government competing against the “private sector”. Will all scientific journalists highlight the major damage to chemistry that this cancellation causes ?
So, Joshua, please keep investigating. Maybe there is a need for another scientific Woodward-Bernstein.
[Added subsequently – my private email favours cock-up rather than conspiracy. But it’s still indefensible to pull grants 6 days before the deadline. And my concern in the previous paragraph still holds – chemoinformatics should have proper public support to support proper science, not languish.]
[1] From Monty Python. In the current case the humour is ironic.

[This is the second grant I have failed to get today. At least the other one – for FP7 – got to the submission stage and was – presumably – reviewed]

This entry was posted in chemistry. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *