Friday, September 28, 2007

The value of failure

It is an interesting life being a postdoc. One big difference from being a graduate student is that you do start to look at life in science a little differently. Contrary to what many people accuse me of having, life as a postdoc isn’t all fun and games, and there are plenty of stresses to cope with. Sure, you have a PhD and aren’t worried about graduation, but there are plenty of other things to worry about. The postdoc time is supposed to be that transition towards a “real job”, and while it used to be a road taken only for people interested in a future in academia, it no longer is so. The past two decades have seen a surge in science spending and growth in departments across universities, but more recent numbers are depressing. While the numbers of postdocs have increased many fold, the number of faculty positions haven’t. So, there are far fewer jobs that a postdoc has to compete for in universities. This leaves industry and startups (small biotech companies) as options for many of us.

Which brings us to the point of this post. Now, particularly in these ultra competitive times, a postdoc requires a substantial amount of high profile work in order to get that coveted faculty appointment. “High profile” can mean a couple of things; the first is work in an extremely “hot” field of research. Something like stem cells (and if you can throw in some more hot topics there, like micro RNA expression regulating ubiquitination that in turn controls histone methylases, all of which controls cell fate, then you are golden). But these areas are obviously ultracompetitive, and often become “hot” when the field has matured a little bit. So, if you enter it at the top of the wave, you are most likely to have to compete with too many people in the field, and may not be able to carve out a successful academic career. However, working in “hot” fields do sometimes give the postdoc the (illusionary) security of believing that even if her work isn’t pathbreaking, a couple of solid publications and mastering of some techniques will be sufficient to land a job in industry.

But there is a second road some postdocs choose to take (and I like to think of myself as one of them), to work on extremely challenging projects that ask questions that aren’t in the mainstream. In this case, all the experimental tools needed may not exist (or may exist but will have to be adapted from something else), but more importantly, the “field” itself does not really exist. The hope is that the hypothesis turns out to be correct, and you will make a discovery that will open up new areas of science. Perhaps this could be described as vertical as opposed to horizontal or incremental research. The problem lies in the possibility that the hypothesis could be completely wrong. Sometimes even if the hypothesis doesn’t hold up, the findings are modest and incremental enough to be publishable. Other times, it is all or nothing, and you could be left with absolutely nothing.

It is sometimes interesting to see how things work in industry. Sometimes with (say) startup companies, the people starting it up have a good idea(s), are extremely competent and take up significant challenges, but for various reasons the company does not take off, and eventually folds. Failure to make the company take off isn’t necessarily viewed negatively. Many people have failed in their attempts to start up companies, but are still highly valued by the industry for their experience and knowledge (of the process of building a company). They remain eminently employable and sometimes also desirable.

Unfortunately, I think life in academia may be a less rewarding for similar situations. Sometimes postdocs (or principal investigators) work incredibly hard on a potentially breakthrough hypothesis for years, only for the results to be unremarkable, or even go against the hypothesis (that could have resulted in a major breakthrough). Usually, you can predict that the person is in store for more hard times in the future, since the lack of any substantial publications means that there’s no chance of being offered a faculty position (or perhaps being denied tenure for a young investigator). It is also unlikely that this person would easily find an industry job, since jobs are relatively few, and industry itself has no clear metrics to measure you by (unlike failing in a biotech startup venture). The way things are set up presently, it is hard to impress someone by showing that you disproved a hypothesis. Most of the postdocs who decide to go down this road are pretty smart, technically competent, and think they can handle hard projects, hope they have some vision (in addition to the vision of their bosses), and are usually aware of the risks. But they are willing to spend the 3-4 years trying. If they fail, then they spend two more years doing something “safe”, which they usually are quite successful in doing, and so manage to get a job in industry after a painfully long postdoc.

So, just using this comparison with industry, can there be any ways to measure the value of a postdoc who has worked on some extremely risky projects which have failed? Is a postdoc who has worked on some extremely challenging projects (which haven’t worked) more or less valuable than a postdoc who has managed a couple of publications, but done so using well established methods and a well established system? Can risk takers become valuable in academia? After all, they’re extremely valuable to investigators who need postdocs willing to go after the most challenging ideas the investigator has, but how can that value carry over to the job market?


Veo Claramente said...

As they say, negative data is data but is impossible to publish.

You're so right on many points, and I think you bring up an interesting idea, is there a way to assign value to the mavericks (and I use that word in the best possible way). I think the first step would be to do away with the "You are invisible unless you publish" mentality.
And on and on, this is developing into a mini-post, so maybe I'll just take off and "comment" in form of a post.

Great post btw,

Sunil said... post on your thoughts, I'll look forward to reading them. The problem with posts like this one is that it is a little hard to articulate all my thoughts, and it is just as hard to respond :-)

Anonymous said...

Hi ,
Fabulous post !
I am a graduate student and dithering over future choices to be made ( not that I am even close to graduating). Your post struck a chord and reminded me of a thread I had come across on bionet in the quest for some obscure protocol :

(thread with get a phd/ world class post-docs)
Its dated, but somehow still pertinent, things dont seem to have changed very much in the decade that followed.


Sunil said...

anonymous....thanks for that link. Not much has changed indeed! And i don't think much will either.

Anonymous said...

I completely agree with your post - I too have taken a project that is "not in the mainstream" and I'm on year 5 of my postdoc. Foretunately I'll have a paper coming out soon, but it's been a rough life. And they say that postdoc-hood is the "best part of a scientist's career" ...

Anonymous said...

I saw you talking about ubiquitin protein, so I thought you might be interested in Science Magazine's current webinar regarding disease treatment:
Science Magazine's Webinar: The Ubiquitin-Proteasome Pathway

Anonymous said...

Mens health drugs

Viagra generic drug is taken by mouth as needed between four hours and one-half hour before sexual activity (about one hour before is most effective). Take only as directed, usually once daily as needed. Sildenafil works along with sexual stimulation to help achieve an erection. Order discount generic or brand Viagra pharmacy online.

Generic CIALIS drug was shown to improve, up to 36 hours after dosing, the ability of men with ED to have a single successful intercourse attempt. CIALIS has not been studied for multiple sexual attempts per dose. Order discount generic or brand Cialis pharmacy online

Generic Levitra drug is taken orally when needed between 4.30 hrs and 30 minutes before sexual activity. Take Levitra only as directed, usually once daily as needed. Levitra works along with sexual stimulation to help achieve an erection. Order discount generic or brand Levitra pharmacy online

Generic Soma drug relaxes muscles and relieves pain and discomfort associated with strains, sprains, spasms or other muscle injuries. Muscle relaxers are for specific muscle injury and pain and should not be used for general body aches and pains. Order discount generic Soma Carisoprodol pharmacy online

Generic Acyclovir drug is used to treat herpes infections of the skin, mouth, mucous membranes, genital herpes, herpes zoster (shingles), and chickenpox in some individuals. This medication does not cure herpes, but relieves the pain and may make the infection clear faster. Order discount generic Acyclovir pharmacy online