Big Mech Boondoggles in an Era of Tight NIH Budgets

September 25, 2013

oh, did I blow the lede on that title?

A letter to the blog points to the NIGMS FeedbackLoop blog and an entry from Jon Lorsch on their “Large Scale Research Initiatives and Centers:. One of these, the Protein Structure Initiative is being readied for decommissioning.

At last week’s National Advisory General Medical Sciences Council meeting, Council members and staff discussed the future of one existing large-scale program, the Protein Structure Initiative (PSI). The Council heard the results of a midpoint evaluation of the PSI’s third 5-year phase, PSI:Biology. The evaluators found that PSI investigators have determined an impressive number of high-quality protein structures and that some of the program’s accomplishments, including methodological ones, could not have been readily achieved through R01-type investigator-initiated grants.

The evaluators concluded that the PSI will reach a point that no longer justifies set-aside funding and, as a result, strongly recommended that NIGMS begin planning the sunset of the PSI, being careful to identify resources developed by the initiative that should be retained for use by the biomedical community.

The blog entry details how the NIGMS only got into Big Mech science during the doubling and generally questions the value of massive projects focused on one topic or theme.

The obvious debate item for this and all ICs that throw money at BigMech science (Centers and Program Projects, assuredly. Sometimes big cooperative agreement U-mechs and contracts as well) is that the more plebeian R01s, investigator initiated and solicited alike, must be sacrificed to pay for larger projects. They have to balance the relative value of a Center or P01 Program Project against the loss of 3-4 individual R01s. Maybe even more of them, depending on size.

First, my disclaimer. I’ve had benefits from BigMech science for a good number of my months of support from the NIH as a PI. Sometimes the bennies amounted to less than an R03s worth of funding and sometimes it has been as large as R21 or small R01 money. I am also at the stage of my career where I not only can be a semi-credible substantial participant in someone’s BigMech but I should be planning for a not-to-distant future in which I try to head one up. And by “should” I mean for the benefit of not just my own lab and career but in a more generational sense for the benefit of my scientific subfields of interest.

I occasionally wring my hands with observation that as the Baby Boomer contingent ages, there are not enough of the following generation to take over the large scale projects that we have at present. This is because both of broader demographic issues (I was born during a particularly low-ebb in total US births) but also because of the over-shadowing, squeezing of my generation of scientists out of academic jobs. So if the larger mechanisms of science funding are to continue, my tiny generation needs to step up.

There’s a final knock-on problem. Because the few of us who managed to transition did so later, and with less assurance of continual funding, we are not ready. Our labs are less stable at this stage of the process. We have not been in them as long at a given age of life, either. We’ve had less time to reach a stage of comfort and (slight) boredom with our own thing that might motivate us to think of the larger picture. Why bother captaining a P01 application that will eventually fund ourselves with R01 money and four of our colleagues when we could just write two or three new projects for our own labs?

I see these pressures. And I see it as a challenge to my generation. Which means a challenge to me personally.

But.

However.

Still.

This all hinges on the assumption and stipulation that Big Mech Projects are a GoodThing.

I am ambivalent.

If you can get one, they are a very GoodThing. That is the tragedy of the Commons answer that governs our existence under the NIH-funded extramural system. I see Big Mechs as barely living up to their promise in the best case scenario. The promise of “the whole is bigger than the parts and SYNERGY”. I’ve been around a BigMech or two and I’m not sure that I believe in the hype. As a general rule, I mean. What the BigMech does in my opinion is provide a very tasty carrot to pull together some investigators around one project. And let a subset (even just one) of those investigators herd the cats a little harder than would otherwise be possible. But there is nothing fancy about the unified project. A group of friendly, collaborating laboratories can do the same thing with individual R01 funding. The extra, add-on Cores that you get with a Center or Big Mech sound good in theory but in my experience may not be much in the way of value added.

And all the monthly Project Meetings and Advisory Board Lunches and other box-ticking stuff? It’s about 10% about the science and 90% about making the next competing renewal go well. IME.

So there is some waste to balance out the theoretical synergistic advantages.

Of course, what is sacrificed is unknown. The Big Mech has to be unified under a common set of Aims or it is dead in the water. So obviously it is throwing a lot of cheddar at one particularly topic or theoretical orientation. That might be wrong or less valuable compared with the hypotheses generated by two or three other groups.

Those other groups will either fail to seek (because those Big Swaaanging GreyBeards have the Center) or failing to secure (ditto) funding for their own, more modest, R01 projects. We know this. What we can never know, is who has the best approach. The funds have to be directed to one strategy or another.

I am a big believer in the underlying structure of a democratic competition of ideas from all comers that underlies the NIH system of funding. Lots of smart people offer their take on the world and collectively we make sure to cover our bases. Great ideas can be recognized no matter where they come from. This is the superior way to get the best of the best ideas at the table.

I am not naive. I know it doesn’t work exclusively this way in practice. But it is the ideal to which we should aspire.

Big Mech projects contradict this ideal. So I am reflexively against them.

Even if it would be in my present best interest to back an expansion of Big Mechs at the NIH ICs of my greatest interest.

Advertisements

5 Responses to “Big Mech Boondoggles in an Era of Tight NIH Budgets”

  1. Dave Says:

    A group of friendly, collaborating laboratories can do the same thing with individual R01 funding. The extra, add-on Cores that you get with a Center or Big Mech sound good in theory but in my experience may not be much in the way of value added

    Exactly. The question is not whether large initiatives produce results (that’s a given), it is whether better (or the same) results could have been achieved using traditional funding mechs for less money and in the same time frame. The report you linked to did make the following comment. Note the “risk-taking” part:

    Some of the accomplishments could not have been readily achieved through R01-type investigator-initiated grants. The establishment of high throughput and membrane centers enabled a concentration of resources for efficiency and risk-taking.

    But isn’t it the optics that count here as well? The NIH announcing big initiatives, and spending millions of dollars on them, while complaining about budget cuts might seem odd to some.

    Like

  2. drugmonkey Says:

    “could not have been readily achieved” is a mouthful of weasel words.

    “would not” would be more defensible but sounds too indefinite, no doubt.

    It would also be fascinating if an analysis held up the 5 just-missed R01s that were passed over in the initial year of funding each competing round for the Big Mechs for comparison of what the alternative would have been.

    Like

  3. Dave Says:

    It would also be fascinating if an analysis held up the 5 just-missed R01s that were passed over in the initial year of funding each competing round for the Big Mechs for comparison of what the alternative would have been.

    Right, but first it is important to quantify how many applications potentially are passed over in favor of Big Mechs. Given the money involved, I suspect it would be a lot more than 5 R01s. How would paylines be affected if Big Mech money were rolled back into the pot, for example? No doubt the NIH would argue…….barely.

    Like

  4. anonymous postdoc Says:

    To what degree, in practice, are P and U mechanisms used to prop up/provide bridge funding to laboratories who are otherwise having a hard time supporting themselves in the current climate? As largesse that can be handed out, kind of like open T32 positions?

    Like

  5. DrugMonkey Says:

    Some. Pilot components are traditional for this. The BigCheez PD can sometimes strongly suggest a component head pick up some pct effort, or play with the Cores. In the longer term, sure the BigMechs can often be used to support a person who is otherwise less competitive for an R. Co-investigators, Core heads, etc. can become a chicken/egg question though…maybe thrown to the wolves some of these ppl would be just fine?

    Like


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: