Today I have the second part of the guest commentary from Zurich's Dr. Theo Wallimann on research funding in Europe. Today he advances his proposal for a new way of funding young scientists:
The definite proof that European Research Programs (such as the FP-6 and FP-7 Framework Programs) are not the sort that basic scientists regard as most useful is the fact that one has to indicate and list so-called "Deliverables". These are research results or products that one wants to or should achieve in the given time period, e.g. being able to express a protein at high levels in bacteria, (Deliverable 1), to purify it to high purity (Deliverable 2) to characterize it by biochemical and biophysical methods (Deliverable 3) and then try to crystallize this protein in order to produce X-ray compatible protein crystals (Deliverable 4).
One then has to provide yearly reports and let the reviewers know whether the goals set were achieved and met in time and whether one could "deliver" as predicted. If one meets one's own prognosis, one is considered a very good scientist who is able to meet one’s Deliverables. In other words, being able to deliver exactly what was predicted is considered good science, at least by the bureaucrats.
But anyone working in the fields of protein crystallization and X-ray crystal structure solving, for example, knows very well that protein crystallization is still an art which often needs a stroke of luck to get good crystals for X-ray studies. This can literally take years, guided first by brute force screening approaches, and if this does not work by intuition and perseverance. All of a sudden, out of the blue, one may be able to grow crystals once, but they sometimes never come again, even if you repeat the experiment under the very same conditions. In those cases you may find out that something subtle has changed, e.g. the battery of the distilled water apparatus was changed and the water quality was thus somewhat different, etc.
I know of an incident where a long, flexible protein should have been crystallized, but many doctoral students and post-docs could not manage to get crystals. After a year or so, a new post-doc came to the lab and started the project from scratch. However, he realized that his predecessors had left many crystallization trials in multi-well plates in the cold room. They must have stayed there for years, some were murky, even greenish and bacteria or algae must have grown in them. The new post-doc could have gotten rid of these murky old plates, but he was smart and clever enough to take his time to look at them.
Lo and behold, he saw crystals in some of them. He opened the micro-chambers with crystals in them, but saved the mother liquor and the buffer drop in which the crystals had grown. Be honest now, how many of you would have done such a thing? But this turned out to be absolutely crucial, for the bacteria or algae that grew in the protein solution drop and mother liquor produced a protease enzyme, which cut the long protein strand somewhere at its most flexible site. The rest of the protein then crystallized. Although this was a somewhat truncated form of the protein in question, the structure of this core could be solved and years later it was the basis for solving the whole protein structure. Why was it so important to save the supernatant and mother liquor? The post-doc cultivated the bacteria or algae, I don't remember exactly, and purified from them the very protease that was cutting the protein at the specific site, such that it then crystallized.
With this tool (the peculiar protease) at hand, he could reproduce what he had seen at first, and was rewarded with protein crystals which otherwise would not have seen at all. I think this is a very nice example of a) serendipity but also of b) a smart experimenter who reacted very cleverly and used foresight in formulating hypothesis that he then could prove to be true.
Would you ever state in a EU funding proposal that you plan to grow protein crystals by letting a protein solution stand around in a messy cold room, in hopes that the right bacteria would grow and nibble the protein’s flexible loop off so that the rest of the protein would crystallize? You would blatantly be considered as totally crazy, I would predict. But this episode took place in the 1980s, not back in Marie Curie’s or Pasteur's time, and similar events can and will happen today.
But such considerations are not a concern of the EU functionaries. They want to see the crystals, especially if you told them you would deliver them in a year or two. This is science on deliverables, as one may call it. But it has nothing to do with daily work in a laboratory. Therefore, as some have pointed out, the administrators should be educated scientist themselves, ones who have worked for a few years in a real laboratory environment. I think this would improve things quite a lot.
My proposition for EU research funding would therefore be: give young PhD investigators (after their post-doctoral training and after meeting various quality standards) no-strings-attached research support for 5 years. In this way they can demonstrate their talent and independence by doing what they like to do, as best they can. If after this time their work stands out, support is then generously extended for another 3-5 years.
After that, the tenure decision has to be made, and those not fulfilling the criteria (to be determined) will leave academia. This would give young people an excellent start-up chance – perhaps then there would be fewer people accumulating in academia who were promised promotions that might be delayed and postponed. (In many of these cases, all of a sudden these researchers are then considered "too old" and fall out of the system completely).
This generous scheme is of course risky, for some money will not be spent the best way it could have been. But on the other hand this will allow the really talented young researchers to thrive and take off for their Nobel PriZe ambitions. So, let’s simplify the granting bureaucracy by being much more generous, while trusting in peoples’ ability to self-organize to meet their challenges and perform. In the end it is not bookkeeping that will count, but the really great and innovative research results that bring humanity a step further along. Why shouldn’t we be prepared to take this risk? I am afraid, though, that my scheme would leave thousands and thousands of desktop offenders unemployed. . .
1. Sili on June 26, 2008 7:57 AM writes...
Where do I sign to get Dr. Wallimann in charge of the EU budget? The entire thing for that matter - he has enough sense to go around.
Permalink to Comment2. processchemist on June 26, 2008 8:38 AM writes...
>But such considerations are not a concern of the EU
>functionaries. They want to see the crystals,
>especially if you told them you would deliver them
>in a year or two. This is science on deliverables,
>as one may call it. But it has nothing to do with
>daily work in a laboratory
For YEARS I heard that academical research needed to be refocused on a "biz" model.
Permalink to CommentIn the research biz if you don't get the milestone payments (for "deliverables") you're in big trouble.
Now that the public funding system uses models mutuated from the research business, it's time for new schemes... you can label this stuff as "evolutionary mechanisms in advanced societies" or "dialectics in history", do as you want.
3. CMC guy on June 26, 2008 10:21 AM writes...
The Deliverables theme does smack of industry situation however they usually are not determined by the people who do the work but are mandated aggressive timelines by managers often out of touch with R&D.
The Open Grant theme, with extension for success, is nice concept although not sure how effective for the spurring innovation and team collaboration it might be. Reducing committee control is generally good but there also needs to be guidance and directions in some form(Scientific Working Groups?) and to certain degree as complete Anarchy might lead to interesting findings that will not be useful (or take extreme work and time or chance connection). In the end for Science to contribute, beyond increase in knowledge, applicable results still must be demonstrated; in academia means papers and/or trained students, for pharma it means good drugs.
Permalink to Comment4. Bruce Hamilton on June 26, 2008 3:00 PM writes...
I'm not sure that, as a tax-payer, I'd like a pool of research money tossed to wildcat researchers to fritter away as they wish. It might be quicker to simply create a new researcher-only lottery.
Is there any indication that fresh post docs will make significant discoveries in the first few years, or do they have to aquire more diverse practical/mental skills to be successful discoverers?.
I agree the current deliverables model sucks, but there should be some value returned to the funding providers, even if it's in the form of negative results that identify research cul-de-sacs along with producing a more experienced researcher.
I wonder if there could be a large shopping list of scientific problems worth investigating, each with some baseline funding for several years, and only new researchers, with at least some of the relevent skills, could apply.
When selection decisions have to be made, vested traditional interests can also appear in the selection process, and that could kill the shopping list concept quickly.
If the only stipulated outcome was publication, whether the experimental programme was successful or not, then at least knowledge and the researcher's abilities, should be advanced.
The research would have to be performed somewhere, and most research institutions would also try to maximise their share of any funds.
Permalink to Comment5. Counterview on June 26, 2008 6:47 PM writes...
If the deliverable methodology is applied as one would to making widgets in a factory I agree it is incompatible with scientific discovery. But what is wrong with setting out *expected* time-frames to deliver results, with the understanding that along the way sometimes "shit happens"? Personally I believe some milestone-driven discipline is a good thing for research. If nothing else it helps focus the mind on what is trying to be achieved in a timely manner. Without such a framework it's easy for projects to become open-ended, blank cheques, frittering away tax-payer's money as Bruce said.
I have no doubt for the early career researcher this is less necessary because the driver for them is to achieve results to further their fledgling careers. For the senior tenured researcher however, it's all too common to see them coast on hobby science long past its use-by date.
Permalink to Comment6. Brooks Moses on June 27, 2008 1:59 AM writes...
I'm reminded of a quote in my quotesfile, that ironically happened to come from here (in an entry dated 2007-11-14), where Derek was quoting Rayleigh's biography of J. J. Thomson:
"If you pay a man a salary for doing research, he and you will want to have something to point to at the end of the year to show that the money has not been wasted. In promising work of the highest class, however, results do not come in this regular fashion, in fact years may pass without any tangible results being obtained, and the position of the paid worker would be very embarrassing and he would naturally take to work on a lower, or at any rate, different plane where he could be sure of getting year by year tangible results which would justify his salary. The position is this: you want this kind of research, but if you pay a man to do it, it will drive him to research of a different kind. The only thing to do is to pay him for doing something else and give him enough leisure to do research for the love of it."
It seems to me that that would be a good way of running research funding. (And, in practice when things work really well, it in fact tends to be. A fair bit of my early Ph.D. research was off on a side project of a large grant that really amounted to this sort of thing on an institutional level.)
Permalink to Comment7. Still Scared of Dinosaurs on June 27, 2008 7:30 AM writes...
The results are unpredictable, but the efforts aren't. If the deliverables were defined in terms that can be predicted - number of related molecules produced, number of animals dosed - then you would have a more reasonable system.
The real key would be that continued funding would be contingent upon adequate analysis and intrepretation of the results whatever they may be. That's a judgment committees can make, at least a lot of the time. I also wonder about some form of auction system where senior scientists have a certain number of points they can allocate to junior scientists according to their estimates of merit. They have time for that, right?
Permalink to Comment8. MTK on June 27, 2008 8:18 AM writes...
Bingo, Still Scared! In terms of deliverables vs. effort.
There is nothing wrong in my opinion with "deliverables" in a research setting as long as researchers allowed latitude in changing the goals and methods of the research as the science and their good judgment leads them.
BTW, from my experience that's exactly how the NIH works. Yes, you need to formulate goals, plans, and milestones in order to obtain the grant, but no you are not beholden to them once you receive the grant. If things change and it is scientifically defensible, it's not an issue.
Permalink to Comment9. emjeff on June 27, 2008 3:23 PM writes...
A fascinationg example of how science is "non-linear". By that I mean the paths can take tortured twists and turns which simply can not be predicted.
There's not a person in the world that wouldn't like to see greater "productivity" in research, but it's not that kind of beast. We have to accept the fact that the Road to Knowledge is a long and twisty road.
Permalink to Comment10. Jonadab on June 29, 2008 7:17 AM writes...
You cannot unconditionally give all the young PhD investigators unlimited no-strings-attached research support for 5 years. There could never be enough money to go around. (Among other things, the number of young PhD investigators would be expected to increase under such conditions.)
Attempting to implement this plan as it is proposed will result in one of two conditions: either the amount of funding each researcher gets would be woefully inadequate for any real research, in which case the researchers would need additional funding from elsewhere, or else the "various quality standards" that the young researchers have to meet would become strings in themselves, and the people who determine whether they are met for any given researcher would become the decision-making body for who gets funding and who does not.
That brings us back to what I said the other day: it's all about who gets to decide who gets the funding and who does not get the funding, i.e., who is on the committee.
Permalink to Comment