# Rat Race

Blogging has taught me many interesting things about academic ecosystems other than my own. I have been particularly fascinated to learn about the inner workings of various disciplines, departments, universities, countries, genders..

Every year, I attend quite a few conferences, visit other universities, advise a lot of students, review and edit 57 million manuscripts and proposals, and collaborate with quite a few other scientists on several other continents, but my view of the academic world would nonetheless be quite limited without blog-input and e-mail from readers.

However, much of what I have learned, although fascinating, has been second-order compared to this:

People in the biomedical sciences seem to suffer a lot more than those of us in just about every other STEM field.

Discuss.

My data: 87% of my blog-related e-mail is from unhappy, bitter, troubled, distraught biomed grad students, postdocs, technicians, and early-career faculty. Others write to me with problems, but these tend to be of the "I'm frustrated with my advisor" sort rather than the "I'm being tortured, abused, deported, sued, and I fear my academic career is over" sort that I routinely get from biomed people.

I specify biomedical rather than the life science in general because, as far as I can tell, the ecologists and botanists and ornithologists and whatnot seem to be reasonably content, or, at least, not more stressed out or bitter than your average chemist, physicist, or engineer. No, it's you people doing the important disease-curing research etc. who really seem to have the most difficult academic lives of all.

Of course there are happy biomed people. I can think of at least 2, maybe 3. And I hasten to admit that I don't really understand much of what I read in some of the biomed blogs, especially all the posts focusing on NIH R2D2 grants or whatever. So maybe I don't know what I'm talking about, but the e-mail data nevertheless indicate that something is going on over there in the biomedical departments.

What could explain this phenomenon? Possibilities include:

- My database is flawed, my assumptions are baseless, my conclusions are wrong. Perhaps there are lots of happy biomed people, including biomed bloggers and blog-readers, but the bitter ones make the biggest impression (and write to me more often than the others). The happy ones have no reason to write, and have other hobbies.

- There are more biomed bloggers and blog-readers and this gives the artificial impression that there are more unhappy biomed people.

- Biomed is a total rat race. Postdocs in my field are respected, paid well (+ benefits), and get good jobs, whereas most biomed postdocs seem to be serfs with bleak futures. Biomed people work in large, fractious groups involving people with huge egos stomping on the peons who do the real work. NIH grants are large, but are not large enough, and are difficult to get. And so on.

Of course, those of us who are not curing cancer are glad that others are working on this, but, if biomed is a difficult and unrewarding career path for many who try to pursue it, can anything be done to fix this? Or is it actually a more exhilarating and rewarding career path than one might think from my e-mail inbox and from semi-casual grazing of the biomed blogs?

Obviously, I have no answer to this question, but perhaps some readers would care to comment?

• DrugMonkey says:

Biomedical folks are whinier?

• NP says:

Biomedical folks (okay, those with PhDs) work side by side with MDs who make 3X more money and are bitter that they didn't have the brilliant plan to get an MD, then go into research (okay, maybe that's just me!).

• JoannaShmoanna says:

Happy biomed postdoc here (who loves reading your blog, but has never written to you) .

• Susan says:

Ed Beck must agree with you as he is in the middle of writing a series for The PostDocs Forum - calling for postdoc unions, specifically in the biomedical field. In his words - Salaries and benefits for Postdoctoral Fellows in biomedical research are extortionate impacting the motivation for US youth to pursue careers in this field.

• ah says:

working at the edge of biomed, I think biomed is a lot more competitive and cut-throat than other areas. There were often cases of PIs telling grad students not to talk to the grad students in the next lab who might steal ideas, of presenting work at a conference only to have it scooped, and of other types of back-stabbing. Plus the stakes are higher when there is potential for a pharmaceutical company to pay big money for your discovery. I am very glad that my field is a bit more relaxed and collegial - we contribute less to curing diseases but have more time to think and do science.

• bouncetwice says:

I absolutely agree - I've observed the same thing. I work in the physical sciences and not in biomed, but absolutely all my friends who do work in biomed have horror stories galore. The stories that are particularly prevalent are those of supervisors who won't let students publish before they've produced enough material to fill 4 or 5 papers (for fear of being scooped), supervisors with proper paranoia about their work, supervisors who don't give credit to their students, reluctance to share any ideas or information with neighbouring labs and (on top of everything else) all the bad things that secrecy and suspicion in the lab provide. In one particular case that I know of, a relatively young researcher was given a big-name biomed professorship and was so unable to cope and so scared of asking for advice that the lab became one big miserable unproductive mess. This was at a very very strong research university (which does almost no teaching), and they were so keen on publicising success that there was no room allowed for anyone to be seen as anything less than be a superstar, even for a short time. This is no way to get science done.

• Tinkering Theorist says:

I think it's their relative closeness to the medical field, where people also seem to be treated badly (at least until after residency). So, if young people are deciding whether to be an MD (maybe MD-PhD) or biomedical research scientist, one isn't so distinctly better than the other, and they both seem more normal.

• qaz says:

Are there non-teaching, research-only soft-money positions in the non-biomed fields? A large part of the biomed field is populated by people whose entire salary and livelihood is dependent on the successful getting of NIH grants. Even those few people in biomed who are in hard-money positions are completely dependent on NIH funding to run their lab. (I know this is not true, for example, of math or physics professors. Even my physicist colleagues who have students have TA-ships to turn back on to fund a graduate student or two. At least at my BigStateResearchU, my biomed friends don't have that.) Given that NIH grants are getting harder and harder to get, this produces major stress.

I suspect that this soft-money NIH-grant-hunt produces second-order insidious problems in the field. It means, for example, that the difference between being a superstar and an also-ran is the difference between survival and being out of science. This is not true, for example, of math and physics professors, who can make a good life out of teaching and small research, if they want. [This doesn't mean they all do. It just means they have more of a safety net than biomed researchers.]

PS. I have a friend who has tenure, but no guaranteed salary. 100% of her salary comes from her grants. In what sense, is this "tenure"?

• GT says:

"PS. I have a friend who has tenure, but no guaranteed salary. 100% of her salary comes from her grants. In what sense, is this “tenure”?"

She is fortunate. Having tenure (despite the possibility of no salary) means that even if her money runs out temporarily, she still gets to stay in the game to continue trying to get more soft money. If she runs out of money to pay her salary, probably her institution will provide her some bridge funding until she receives her next grant.

As a non TT soft money researcher (I was a research professor), I did not have any bridge funding to tie me over until my next grant. not just my salary but my very existence depended on soft money. When my money ran out (grants didn't get renewed, new grants didn't get funded IN TIME), I not only lost my salary I also had to leave science. I lost my entire livelihood. Game over.

At least for your tenured friend, being money-less would mean just that she has temporarily lost her salary. It doesn't mean she has to leave her position, her lab, her livelihood, everything she has ever worked for. She gets to stay in the game until she wins again.

• Carla says:

Fortunate?! I've come to the conclusion that none of these awful conditions in STEM fields (and academics in the humanities might be even worse) will ever change as long as there is a steady supply of people who (i) would think of a situation like that for a highly qualified professional as "fortunate", and (ii) would want to "stay in the game" under those conditions.

• Deb says:

I agree with Carla. Smart, highly educated and qualified people need to pick themselves up off the floor so they can no longer be walked on before this situation might can.

• A. Nony Mouse says:

As a biomed postdoc, I'm certainly experiencing a lot of angst about my career. I don't know whether its that my situation is worse than non-biomed STEM, or if its just me and my personality. However, it does seem to be a pattern, and I can't see how biomed would self-select for 'whinier'.

A lot of people blame the NIH budget doubling (1998-2003). (e.g. here) A lot of that money has gone to build new/larger medical and graduate school programs, increasing the number of graduate students moving through the systems. It's as if, during the ramp up, people got too used to that growth rate and acted like it would continue indefinitely rather than leveling off again. Originally, the NIH budget was supposed to keep increasing more slowly, but still above inflation. Given the economy and the budget climate, that hasn't happened.

Are salaries lower? Perhaps. The prestigious, but generally regarded as underfunded Ruth L. Kirschstein fellowship pays $38k-48k for postdocs with 0-5 years experience. Some advisors use that as a benchmark (and won't supplement the salaries of postdocs who receive them). Other advisors pay more. I certainly make more. I benefit, I think, from being in a hot sub-field, and one that is computationally intensive. Stipends have to be higher in order to recruit and keep people with the necessary computer science training. • I am totally jealous that, as a biomed postdoc, you are making more than the fellowhship standard. In my university, it is actually unusual for a biomed postdoc to make that level. Typical compensation is closer to$35-45K for 0-5 years experience. We also only get a portion of the benefits package available to faculty/staff: health/vision/dental, but no retirement or family leave. And we ARE the ones 'curing cancer'.

Recently, however I learned that the majority of my labmates from China are making far less even than I am. In the range of $12-20K per year. Does anyone have any insight into how that works? I do have to say though, there are several new biomed research buildings on campus... • Simon says: Well for a start some basic investigation suggests the cost of living in China is at most 1/2 the cost of living in the US, so you have to keep that in mind when you're comparing... • Carol says: but they're in the United States doing this postdoc though. I hope they aren't in California or Boston trying to pay rent. • anonymous says: First off, I wouldn't be surprised if, in fact roughtly 87% of grads/postdocs ARE biomed. This is simply because BioMed has a HUUUUUUUUUGE glut of PhD's due to the larger amounts of funding available from NIH, AHA, and the American Cancer Society compared to other agencies. So even if biomed PIs only outnumber other scientists 1.5:1, their "trainees" will outnumber everyone else 10:1. That said, I think that for this same reason, BioMed HAS become a huge rat race. Take for example the fate of postdocs. In every other field of science, post docs are 2-4 years unless there are unforseen problems/life issues. In NIH funded labs the AVERAGE is now roughly 5 years. PP and DM and the other biomed PI bloggers would have you believe "Well! BioMed Science is SOOooo much more complicated and SOOooo much more sciency - you NEED 6 years post grad just to get going!" This is bullshit. The reason that postdocs are longer in BioMed is for one reason - PIs can get away with it because of competition amongst the glut of PhDs that are out there for the real jobs. The glut exists because of bizarre funding structures that allow super-labs with 10+ grants and no labor-practices oversight. You get lovely situations like postdocs and grads from other countries being told they must work 80 hours per week or they will be deported. Or even better hiring them as techs after graduation and paying hourly but still demanding 80 hr/week. Yay. Anyway rant over. BioMed sucks. Kids, pick a different scientific specialty. You're not going to cure cancer and your PI doesn't really think your research will anyway - s/he just says this to get more grant money. • drugmonkey says: DM...would have you believe “Well! BioMed Science is SOOooo much more complicated and SOOooo much more sciency – you NEED 6 years post grad just to get going!” That is a scurrilous lie. Please show where I have ever said such a thing in anything other than the descriptive sense or as a criticism of what others seem to believe. Otherwise, pistols at DAWN, muppethugger! • Grumpy Lurker says: Well! BioMed Science is SOOooo much more complicated and SOOooo much more sciency – you NEED 6 years post grad just to get going!” Not sure what DM said or not, but this attitude oozes from a large number of biomed blogs. The aura of perceived superiority. That everything is so much more grandiose in the biomed -- the money, the science, and let's not forget -- the suffering! Oh, such deep, deep suffering! Biomed fields are the fuckin' French cinema of science. Gimme a fuckin' break. I hate plants and animals and would never go anywhere near biomed research, yet I know more about the NIH funding (fuckin' R01, R21,K and F whatever) than I ever wanted to. I am quite starved actually for more blogosphere discussion about non-NIH funding and the different obstacles that physical scientists take en route tenure. Btw, I loved my PhD and my postdoc experiences (I am a 2nd-year TT prof now); both were in experimental physics, in labs with 6-8 people, of whom 1 or 2 postdocs. The money was never so abundant that any one person was disposable; the PIs made darn sure everyone published well and got enough exposure at conferences. Maybe if there was less money overall floating around for biomed research and fewer fellowships for postdocs (so each postdoc would have to be directly supported by the PI), the biomed PIs wouldn't be such asses to their trainees. • drugmonkey says: Given that all this nonsense you are blathering is not at all consistent with our blogging at DM, save the R2D2 bit, one wonders whether you have any evidence at all or are instead reflecting a sense of inadequacy generated elsewhere. (don't get me wrong, biomedical topics ARE objectively better than fizzycs. ) • Point taken. I'll try to write more about NSF funding. • I have never said this either, you fucken liar. • Ash Ash says: CPP, more than once you going off on a postdoc going on the job market that they cannot possibly be ready for PI-dom after x many years. Without knowing the specifics of their career. This default attitude definitely has a bit of that comment up there in it. • Curt F. says: I think the key to having miserable grad students and post-docs in a scientific field is: (1) Have the primary funding agency in your field double its budget over the space of ten years or so. From 1980 to 2006, the NIH budget in constant dollars went from about 7 to 27 billion dollars, a 3.8 fold increase. (2) Make sure that the budget doubling is used to vastly expand graduate enrollment in your department. In 1981, nationwide, about 5800 life sciences doctorates were awarded each year. In 2006, that number was closer to 9800 per year . (3) Also make sure that the number of PIs funded by your field's funding agency increases at a much lower rate than the actual budget. NIH had about 9000 PIs in 1980 and about 21000 now, a 2.3 fold increase. That means that there are 65% more dollars per PI now than in 1980. (4) Make sure it is clear to students that data generation (i.e. lab work) is their primary function. It is better if they realize that the lab work shall be in pursuit of one of your hypotheses, not theirs, and also if they know that you are pretty convinced of your hypothesis. If one of your underlings finds contrary evidence, it probably means they are not very good at doing experiments. (5) Also, make sure that the lab work your underlings are expected to do involves tedious procedures that involve lots of manual pipetting. Make sure that these procedures, which they must invest massive amounts of time in learning and executing, have no even remotely similar counterpart used in any other field of science or business sector. That way, your students will have no in demand skills outside of your field. They'll have little choice but to stay working for you! (4) is optional. Even if you treat your underlings with respect and dignity, factors (1) - (3) and (5) will still conspire to make them miserable. • ScienceGeek says: NIH paylines are and have been much lower than the NIH proposal review system requires for fairness. The NIH funding system functions well with paylines between 15 and 25%. With paylines below 10% year after year, who gets funded boils down to luck and whether your friends happen to be on study section. PIs are frustrated by the capriciousness of grant review and, because they are human, can only shield their postdocs and grad students so much. The job prospects for postdocs are not good in academic research, so being underpaid and receiving poor or no benefits is no longer an appropriate trade-off. While so far, I am doing fine in the NIH meat grinder, I see body parts and fear in every direction. • Applied Physics Prof says: NSF finding rate has been 10-12% in most physical sciences for years. And no study sections where you can revise a grant and it goes to the same people, like what NIH has -- each submission is a crapshoot. DOE is at 5%. I wouldn't say there's much fairness there, so that's not a reason alone for the trend FSP noticed. Physical scientists have it worse in terms of funding, yet the bulk of complaining comes from the biomed. • Dr. O says: Do physical scientists still get paid if their NSF or DOE grant doesn't get funded? • Applied Physics Prof says: Faculty have 9-month salaries paid by 'hard money' (institutional funds) which is essentially compensation for teaching; for faculty, no grants = no summer salary (unless you teach over the summer too). For everyone who is on 'soft money' (a.k.a. from grants), which is pretty much everyone but the faculty, it's as bad as you can imagine: grad students and postdocs have to be let go if no funding comes through. Staff scientists and technicians are less common than in the biomed (except in very big labs) and they too get let go if funding doesn't come through. • ScienceGeek says: Physical Science isn't dealing with severe downsizing right now because funding has been more or less steady for years. That makes a big difference. • A. Nony Mouse says: Something I meant to add in my first comment, but forgot until after I hit post: The length of time to 'independence' is long, and getting longer. A major career milestone for anyone on the tenure track is the NIH 'R01' grant - these are large-ish grants, the kind you can sustain a (small) lab with. Without one (or equivalent), you'll have issues getting tenure; with one, if you don't get tenure it wasn't because of funding. People with larger empires generally have multiple R01's, in addition to other grants. (Oversimplifying, but that's the gist.) The average age for first-time R01 grantees: 1970 - 35.3 1980 - 36.8 1990 - 39 2000 - 42 2007 - 42.6 I don't really know if this is different from non-biomed STEM, but it doesn't seem healthy for the field. • m says: Anecdotal evidence here: I'm in ecology and evolution, but while I was applying to grad school, I ended up interviewing at one university's biomed department. At the recruiting weekend, I was the only interviewee not wearing a suit (I'd actually come straight from the field -- they were lucky I didn't have holes in my clothes!). The conversation between the interviewees tended to follow the narrow path of: what other schools did you apply to? you got in there!?! what were your GRE scores? Every once and a while they took a break from this to brag about their undergrad projects. It seemed like a foreign country: the rest of my recruiting weekends were basically excuses to eat the local food, drink the local drinks, and try to figure out which of the other interviewees would be your friends next year. That being said, I'm also a teensy bit jealous of those in biomed because that they get to, you know, cure diseases. • Bagelsan says: God knows studying ecology could never contribute to human health improvements! And the curing of diseases doesn't require an understanding of evolution at all, really. ;p • m says: Ok, you have a point 🙂 • Sheesh, those interviewees sound awful. I interviewed with programs that generally had a mixture of biomed and basic bio (though for the sake of this discussion I'm not sure if these are different?) research and I didn't hear anyone ask what other people's GRE scores were. Who would want to have that type of attitude in your peer group? On the other hand, my own research is very unlikely to ever contribute directly to curing human disease, so the medically-inclined bio people have that on me as well. • Carol says: I had a very similar experience when I got into an integrative biology dept at one school, and a genetics dept at another school. The genetics school spent a lot of money trying to convince us to go there by taking us on tours to wineries, brought out biotech company reps to encourage us to spend less than 4 yrs on the phd and motivate us to think ahead, and the cohort of other accepted students were very much like the description above • moom says: Maybe they are wondering why they didn't train for medical practice instead and make lots of money. • quasihumanist says: I am a pure mathematician on the job market, and I always thought that on that front we had it worse than all the scientists except astronomers. Some jobs I applied for attracted over 800 applicants. Although a small number of these applicants have jobs and a few are unqualified, most are grad students or postdocs who need a job next year. There are about 150 permanent positions and maybe about 50 temporary positions for which I can pretend to be qualified. (I am excluding the ones that specify a subfield I am not even remotely in; my PhD is too long ago for most postdoc positions.) (What does everyone who doesn't get a job do? Some adjunct, as in the humanities. Some find positions in their home countries in the developing world. The rest drop out of the field and do who knows what.) That being said, there are some reasons that nasty competitive behavior doesn't happen much in mathematics. 1) We don't need a lot of funding. In fact, most of these jobs are primarily teaching positions which have no expectation for writing research grants. In any case, mathematical research is cheap. 2) There are not a few really important problems that everyone aims towards. Instead, there are far more problems than there are researchers. One gets ahead in the mathematical world not by solving problems deemed interesting before you start, but rather by convincing other people that the problems you are solving are actually interesting. Therefore it pays off to be a nice person. • aaa says: I am not a mathematician, but in a part of computer science close to math. In my discipline most grad students and postdocs who do not get academic jobs end up working in Wall Street. I can't say they are any worse for it. 🙂 • Jen says: Perhaps it has to do with the proximity to the medical industry. That has always struck me as more of a rat race and much more "Big Science" than other areas. • Pascale says: Offhand, I would say it was the relative glut of money compared to other sciency fields and the failure of the funding pie to grow with the number of folks we have trained. But I am a biomedical researcher reinventing myself into a non-PI, so my views may be jaded. • Psyc Girl says: I actually think Psychology is very similar in this problem, particularly because it shares a lot of the characteristics people here have outlined (very competitive, only a small number of grants, research structure is very much a hierarchy with grad student minions at the bottom, more PhDs than jobs, etc etc etc). • neurowoman says: I think there may be more pain in the Biomed field because the pyramid structure is much 'wider' - for every PI, you have several postdocs, more students, lab techs, plus a bevy of animal support and other staff to support. You can't do competitive research without a lot of money (you can easily burn through$5k a month just for reagents and disposable supplies for a small-moderate lab). Research is labor intensive, thus all the staff. There are way more trainees than permanent positions, so the competition for those permanent positions is more intense. Once you've spent 20 years becoming a specialist in your field, the non-academic job options are slim. Engineering postdocs can always walk away and get a good paying job, biomed postdocs are out of luck (many, such as physiologists like myself, just aren't qualified for drug industry jobs). It's a brutal up-or-out system with the odds stacked against you, while you work hard for decades for low pay. Ecologists who are lucky enough to get a faculty position at a PUI can pull in small grants to do small research; a biomedical researcher who needs a 2-photon confocal microscope and $500/vial reagents can't do the research they were trained to do at a PUI - they are out of the 'game'. • jen says: "anonymous says: March 22, 2011 at 08:06 First off, I wouldn’t be surprised if, in fact roughtly 87% of grads/postdocs ARE biomed." That's exactly what I thought when I read FSP's post. Does anyone know the statistic for this? That being said, I'm a new Asst. Prof in Biomed. I love it. No, it's not easy. Yes, I've faced discrimination and low funding levels and frustrations. But I love the science, and I've been lucky to work with some fantastic people (and, yes, some less-than-fantastic ones). I somehow got extraordinarily lucky enough to land both a grant and a good job, so maybe that unfairly colors my viewpoint but: I'm a very happy biomed person. And I've read FSP's blog everyday for years and I have never e-mailed you. So maybe (a) a disproportionate number of postdocs are biomed and (b) a disproportionate number of people who e-mail you are acutely upset. After all, I don't imagine you get lots of e-mails that say, "Just FYI, this biomed field I'm in is tough - but I really love my job." • roadnottaken says: According to the 2009 'Survey of Earned Doctorates' people earning STEM PhDs are: 32.4% - Life sciences 23.7% - Physical sciences 22.3% - Social sciences 21.7% - Engineering So, yes. Life science research is over-represented, but 87% is way off. FWIW - I'm a happy biology post-doc. Academic job prospects seem OK and post-doc life is not that bad. • jen says: Hmm, those numbers are helpful - but - those numbers are for PhD's awarded in the US, right? That is an indicator of the number of postdocs, but only partly - BioMed's are more likely to do a postdoc (I think?), and more likely to do a longer postdoc (I think?), and there are a lot of non-US PhD's who are postdocs in the US... • I'm a biology graduate who chose the Eco-Evo rather than the Biomed route. PhDs in Biomed pay roughly 25% higher than Eco-Evo, with more time available for research. Many more RAs than TAs. My impression is, for this reason, it's more competitive at the recruitment level. And it attracts more "competitive" candidates (and I mean that in a personality-trait way). From there up, I don't know what happens to the actual payscale, but it seems like the job market gets tougher and the payback stops being as much better for the extra stress. • chall says: Personally, my experience of the biomedical postdocing is that since there are so many post docs in one lab (or field), and everyone is looking for "the paper", it's hard to feel that you are among friends but rather competitors.... not to mention that when you are looking for research positions afterwards, it's been a lot of MDs who do research but can get salary by working as a physcian who are more attractive. That said, I guess it is partly "MDs vs PhDs and their research training"? It was different when I was in a smaller field (non biomedical) as a graduate student, since we did collaborate with several of the other groups within the field. Considering there were like 5 groups in the world, and we all knew eachother, it was an huge difference. Not saying we didn't "compete" but there was a fair number of sharing ideas, approaches and trying to make the best research and publish. • Alex says: I'm a physicist who works on things that are useful in biology, so I meet biomed and non-biomed people. Biomed does seem weirder. For starters, they refer to their advisor as their "PI", while I always referred to him as "my advisor" or "my professor" or often just "Dave." Being a professor myself now, I am a PI on a grant, but I am an advisor to a student. Oh, and every biomed lab seems to have a tech. Physics labs rarely have such people, unless it's a center with a mega grant or something. I'm sure there's an exception, but it's rare. Also, they don't teach much. Even at R1 schools, it seems like most physics faculty teach a full course every term. They may have TAs for the course, and they may not be responsible for the labs, but they have to lecture 2-3 days/week, every week. The only exceptions that I know of tend to be people with huge mega-grants to run big centers. Biomed faculty, OTOH, seem like they often just teach a portion of a course, and not even every term. I realize that there are perfectly good reasons for this (e.g. they bring in more money) but culturally it makes them even more different from the rest of the university. Also, biomed folks have a weird idea of what interdisciplinary is. In physics land, interdisciplinary means that you have a biologist and a physicist on the project. If we had, say, a theoretical physicist and an experimental physicist, or two experimentalists from different areas, we would call it collaborative but not interdisciplinary. In biomed land, it seems that if you have a structural biologist and a physiologist working together, that is interdisciplinary. (I don't know if a project with a herpetologist and an ornithologist would be considered interdisciplinary, or if they'd call it interdisciplinary if an expert on trees collaborated with an expert on fungus.) All of this just adds up to a really weird culture, compared with the rest of science. As to qaz's question: Are there non-teaching, research-only soft-money positions in the non-biomed fields? A large part of the biomed field is populated by people whose entire salary and livelihood is dependent on the successful getting of NIH grants. I've seen soft-money physical scientists, but not many. A big department might have a couple of them running around, but they would be rarities. • Anonymous says: What?! There are tons of soft-money research scientists in the physical sciences!! • Gecko says: Yes, there are a lot of them in the earth and ocean sciences, for example. Big institutes like Woods Hole, Scripps, Lamont-Doherty Earth Observatory (Columbia Univ) are crawling with them, and they are also quite common at large research universities, both public and private. • TUL says: I'm a soft money physical scientist (I have a joint appointment with a university and a government lab) as more many of my peers who were postdocs in physical sciences and have since been converted to staff scientists but on soft money like myself. Maybe we are different because we did not enter the academic job market for tenure track professorships, our career objective was to remain as staff scientists in government labs. The point is, even government labs are cutting back on hiring scientists, so those who finish postdoc stints have to either leave research (go to industry, or change career, or become stay at home moms) or cobble together soft money positions in order to remain active scientists. In this career approach there is zero job stability and the chance for career advancement is extremely vague (it depends entirely on the details of your soft money position and how it was put together...some people manage to become PIs with their own labs and staff and job stability more than 2 years at a time while others seem to be stuck in postdoc-land...) so I dont' recommend it unless you are fortunate enough to be able to put together something that works for you. Still, my experience is that there are quite a few physical scientists, long past the postdoctoral stage, employed in entirely soft money positions. • Alex says: OK, I was mostly thinking about university physics departments. I see some aged postdocs in soft money research scientist positions, but I don't usually see many "Research Professors" (i.e. 100% soft money faculty) in physics departments, while med schools are chock full of them. In fact, as I understand it, even "tenure" in a basic science department of a medical school is really soft money, since they have to raise their salary from grants. • TUL says: "'I see some aged postdocs in soft money research scientist positions, but I don’t usually see many “Research Professors” (i.e. 100% soft money faculty) in physics departments" probably because many (or maybe most?) university departments will not give lab space to non tenure track research faculty unless these people "already have" their own grants or equipment such as from a previous job/position or if they are fortunate enough to have an established yet charitable PI on their side who would let them share their lab space (without treating them as postdoc underlings) or let them be co-PIs on new grants. Otherwise, the funding rate already being dismal, it's even harder to get grants if you don't have any institutional support. Yet your institution won't support you if you don't have grants. Chicken and egg problem. Thus, soft money scientists can only seem to be employed as lab staff on some rich PI's grants or in university centers. This means being a staff scientist, not a PI. Just my observations in my neck of the woods. • yolio says: I am a biologist of the evo/eco variety. But I would say that structural biology and physiology are wildly different disciplines, and this definitely qualifies as interdisciplinary. It is not at all comparable to experimentalists from two different areas of physics. Biology has a lot more breadth than physics. • GMP says: Biology has a lot more breadth than physics. How about we all refrain from making sweeping statements (especially derisive ones) about fields we are not well acquainted with, OK? I am sure there is plenty of breadth in biology, but I assure you there is also plenty of breadth in physics, chemistry, math, computer science, geology, etc. • Alex says: Also, neurowoman makes a good point: In most non-biomed fields, a PUI job is an option. Yes, those jobs are still competitive (100+ applications/position) but at the end of the day I'd rather be able to apply to 2 types of job than just 1 type of job. But many biomed grad students don't even have TA experience. I won't pretend that being a TA is enough to get a job at a PUI these days, but it's a start. A lot of biomed grad students seem to lack that experience, and also have all of the expensive research needs that neurowoman cites, so PUI jobs are harder for them to get. Having a bunch of postdocs who are focused not just on academia, but on a particular slice of academia, must make it even more competitive. • Anon says: Uhh, there's more women in biomed than in some other STEM fields. We (i.e. women) just whine more, that's all. • Bagelsan says: Probably the entire field has all our periods synced up, too. :p • fubarator says: Could there be a pink-collar effect in fields with increasing numbers of women, where women at the entry level are viewed more as temporary workers, and less as successors to the big boss? Over the long term that could make conditions tougher for all grad student and postdocs in those fields. NIH funding doubling rollercoaster probably has something to do with it too. This reminds me of the 48 year bamboo flowering cycle, "followed by a plague of rats and famine." http://en.wikipedia.org/wiki/Melocanna Rat race indeed! • Alex says: I don’t really understand much of what I read in some of the biomed blogs, especially all the posts focusing on NIH R2D2 grants But those are the most awesome ones--they go to computational biologists working on ways to hack the controls of the Death Star! Also, the use of a Star Wars joke proves that FSP is definitely a physical scientist. • msphd says: Brava! Yes! Maybe we can understand each other better from now on. re: postdoc statistics: see for example this report from 1999 I don't understand this table, the numbers don't really add up (?), I feel like there's a column or legend missing with more labels? And I can't find aggregated data from ~2002 on. Maybe there hasn't been another report recently, or I'm forgetting which agency did the latest one. Top three reasons I would use to explain why biomed is so much worse: 1. Experiments take too long and we publish much fatter papers but only infrequently (1 every 2-3 years is common). Longer time scale => accumulation of trainees => increasing competition => increasing expectations => longer time scale with a moving finish line => accumulation of bitter trainees who can't find jobs even when they've met the unstated requirements... Also, big fat papers are bad because it means fewer people will actually read the paper (as opposed to chemistry where many papers are 1-2 pages, or engineering where papers might be 2-4 pages). So we're even more dependent on reviewers at Cell, etc. to determine our fate. And the longer it takes for a paper to be considered finished, the longer the PI can hang that over your head, because if you leave before it comes out, you have nothing to show for all the work you've done, and you won't be able to get a job... 2. Pyramid scheme. Inflated requirements for publication require large teams, but not all team-members will ever get to move up the ladder, and the lack of sufficient alternative slots in biotech coupled with a large(r?) influx of foreign trainees leads to a glut of overtrained, underpaid researchers with nowhere to go, which leads to abuse because PIs know they can exploit the pyramid scheme and still have plenty of willing replacement workers to choose from. 3. NIH doubling coupled with public naivete. The public perception that there must be jobs in biomed, fueled by this assumption that because diseases are bad, we must be funding research on them! Right? Several times a year I meet a parent whose high-school age kid is interested in doing biomed research. In *every* case the parent is convinced this is a good plan, very prestigious, and they assume that because it is fairly difficult there must not be that many people who can do it, right? Better than most other career paths? They never believe me when I tell them there are not nearly enough jobs to go around and that it's kind of a risky dead-end road to take. Sure, it's scenic, but it ends overlooking a very steep cliff. Also, since I've been unemployed, I've met a lot of otherwise educated people who had no idea biomed had these problems, or what it means if I say that a lot of big pharma companies are in trouble because their blockbuster drugs are going off-patent in the next couple of years. Sure, they might be thrilled to get Allegra OTC or a generic cetirizine, but they don't extrapolate to what that means for jobs in the biomed sector. • Dr. Moose says: as opposed to chemistry where many papers are 1-2 pages.. It's interesting how little we know about each other's fields and yet are willing to make these statements. I publish ~10 papers/year, and this puts me in the top-middle for publication rate in my department at an R1. My 10 most recent papers ranged in length from 4-40 pages (the 40-page one was unusual). It looks like most of them hover at 20 +/- 3. • Engineer says: Also, big fat papers are bad because it means fewer people will actually read the paper (as opposed to chemistry where many papers are 1-2 pages, or engineering where papers might be 2-4 pages). So we’re even more dependent on reviewers at Cell, etc. to determine our fate. Your statement about engineering is false. I am an engineer, and my conference papers are 10-12 pages, and my journal papers (which include more details) are anywhere between 20-40 pages. • dr smith says: I'm a BioMed assistant prof, and I agree with most of the points above. Lack of funding, "pyramid style" super labs, and hyper-competitive personalities can definitely combine into a storm of miserableness for a student or postdoc caught in the middle. Another issue is the leadership structure. Most BigName Biomed programs are associated with BigName MedicalCenters. The politics of these things are ridiculous. BigName Executives get hired to run the show, re-organize everything to suit their master plan, then leave for an even bigger name job before anything gets implemented. There is a lot of talk about being cutting-edge and innovative, but only if you're raking in money from patients! The peons (most of the faculty) just keep their heads down and try not to get RIF'ed in the latest round of budget cuts. An appointment in a department that has both clinical and "basic science" faculty can also be tricky. At our institution, state funds for teaching are cut more and more every year, and the clinicians are exhorted to find even more "billable hours". So the MDs need to be working the hospitals while they're also supposed to be training residents, and the PhDs aren't getting paid for teaching. PhDs are threatened with termination if they don't pay their own way via NIH grant. Teaching is expected, but not compensated much, if at all. And grad education is hardly subsidized at all (no TAs or institutional RAs). It's no wonder grad students find themselves chained to the bench. Classwork is just taking time away from data generation. However, I think there are still small labs doing good science. I'm starting my own right now, and so far I'm in halfway decent shape. I do teach, as well, and some of my salary comes from that. The advantage to being in BioMed is that I get access to honest-to-goodness patients, and I get to collaborate with MDs and work on more immediate sorts of biomedical problems. If you get lucky, this can make a nice career if you make a name for yourself in the disease-specific field. But in 3 years I could be writing my own complaint e-mail or out of a job, so take that for what it's worth.... • GMP says: as opposed to chemistry where many papers are 1-2 pages, or engineering where papers might be 2-4 pages I'm sorry, but I know of no physical science field that routinely publishes 2-page papers. If anything, those are called extended conference abstracts in my physical science field. Letters are typically 3-6 pages, depending on the journal, comprehensive papers can be as long as you like (e.g. I have had an 18-page paper as a single author). Physical scientists also have Nature and Nature Progeny papers that take years to get the whole story and write it up. The difference is that most groups don't expect *all* of their papers to be GlamourMagz and they also publish comprehensive papers in society-level journals, so it's not just 1 big-ass paper for the whole lab. I guess a good question is why is it permitted/expected in the biomed to have huge labs with huge amounts of money and only produce big papers so infrequently, whereas in physical science areas such as chemistry and many branches of engineering you are expected to have a much higher output rate (infrequent GlamourMags won't cut it). Obviously it's a discipline cultural issue, and I'm curious how it came about. • jen says: @GMP...in my biomed field there is no expectation that "all" pubs be in Glamour Mags. I've worked in BioMed at Big-Name Ivy League Schools and I think people *hope* for S/N/C, but really...with the exception of a Rock Star here and there, if that happens 1-2 times in your career you are lucky. (Unless you are a crystallographer...) As for papers being much more infrequent, that much is totally true - I think part of it may be because of how long it takes to do certain types of experiments - especially animal experiments. In my biomed subfield, it's kind of expected that you will use a KO mouse to prove that your protein plays an in vivo role. To make a KO mouse is a year, to backcross it is another 2 years, and each time you want to get animals to (gasp!) actually do an experiment it's 6 weeks minimum (3 weeks gestation, 3 weeks until weaning)...you have to hope you have animals of the right genotype born, and then the experiment itself may involve putting them on a diet/treating them with a drug/etc. for weeks-months-years. I'm NOT saying that we are so so awesomez - I'm just saying, "Damn - it takes a long time." After all that (which may be only be 1-2 figures in the paper, but represents 3-4 years of work and$100K's of grant money) yeah, there is pressure to get your One Paper into the Biggest Journal Possible. For sure, there are biomed experiments that can be done in days and they are done all the time. But I think that the continued push for transgenic animal models in order for any story to be considered 'complete' has made it hard to get out a solid paper in a short amount of time.

My PhD advisor did his PhD before the "KO mouse" thing was a requirement to show something was important, and he had 5 first-author papers his 4 year PhD. He tells me this was common then. (I had 2 first-author papers in 5 years.) So I really think the "infrequent big paper" phenomenon is rather recent, and I blame the mice. 🙂 Whaddya think?

• Alex says:

I’m sorry, but I know of no physical science field that routinely publishes 2-page papers.

I think it's JACS that has 2 page Communications....with 30 pages of supplementary materials.

• becca says:

Statistically, across the country, the number one source of job related stress is... one's immediate supervisor.
Maybe biomed types make lousier managers? They might have a lot more trainees (particularly if they are any good as mentors- it's a bit of the Peter Principle- you keep adding graduate students/postdocs until your competency as a mentor is minimal). They spend less time teaching, and so may resent any role in which their leadership has to emphasize the *other* person's growth. In addition, many PI's see themselves as managers/bosses of small semi-independent companies, because of all the fundraising soft-money faculty are expected to do. The PI vs. advisor thing is more than terminology- it reflects a default lack of a mentorship based relationship with one's PhD advisor

It may also be worth pointing out that many big name medical centers are *much* more likely to be profit-driven enterprises than universities (we're not in 501c(3) Kansas anymore here, folks). The perplexing peculiarities that sometimes result from being a 'college of medicine' employee vs. a 'medical center' employee are truly strange. Sometimes I feel like the more applied science one is doing, and the more clinical collaborators one has, the more the for-profit driven model seeps into everything.

• Worm Pilot says:

Wait what?!? Scientists outside of Biomed actually LIKE their work??? And get paid more than the base minimum for the NRSA over at NIH???? Damn...I picked the wrong field. Sigh. Wish I had read this before 5 1/2 years of grad school and 2 years as a postdoc.

• Wow...I was a reasonably content life/biomedical scientist until I read this post and the ensuing comments...hmmm

I tend to think that an underappreciated problem (and all the one listed above are good ones) is the duration of the PhD programs. In a sense, they're too short... Many trainees take 4-5 years to finish, but some programs fill the first 2-3 years of course work/rotations (especially ones at smaller universities, in my experience). This is beneficial, but doesn't really provide the opportunities to real develop your scientific talents. This leaves you with a situation with (a) mediocre PhDs who are now unemployable, (b) Good people without the skill set needed to be employable. Hence, the glut. In my imaginary perfect training program, it might make sense to have the PhD programs longer, which would have an advantage of (a) Slowing the rate of new post-docs/shortening post-doc, (b) scaring off a few more people from the field, (c) Encouraging unhappy students to drop out midway (i.e., if you aren't happy 3 years in, you might slog it out for 2 years more, but not 5 years more)

Also, i think the drive to publish a big paper in S/N/C is caused by the high pressured job market, not the other way around... Its one of the few markers to stand yourself out from the 200 CVs on the job pile...

• Dr. O says:

In my imaginary perfect training program, it might make sense to have the PhD programs take longer...

This is a terrible idea. Pushing back the time when scientists can get a *real* job would only select against people who want to have a life outside of science...kind of not the goal when trying to select for *happier* biomed scientists. Five years is plenty of time to develop the critical thinking skills required of a PhD, even with coursework and rotations. Besides, time spent in a grad program isn't (usually) the determining factor for when a candidate defends his/her dissertation, so there is no real way to extend the time to graduation.

In my hypothetical program, the extra PhD time comes with shorter theoretical post-doc time, so the time to start a "real" job is the same.

Five years is definitely not plenty of time to develop the critical thinking skills, perhaps for some, but not for many. I would probably guess the old farts (PP, DM, chime in?) would suggest that critical thinking is not a "have it or not" but a constantly improving trait, even after starting TT?

Any grad student or post doc or TT can have a life outside of science... anyone who says otherwise is either working for a tyrant, or trying to convince themselves that quantity of hours worked is more important than quality.

Extending graduate program durations as part of a concerted effort would be possible, but it would require (a) departments to be on board and require more work in a thesis, and (b) funding agencies/universities to provide funding for more than the standard 4-6 years. I know plenty of humanities/social sci. PhD students who are in year 7-8 and are not being pushed out the door, so this model clearly exists in other disciplines...

• Alex says:

I know plenty of humanities/social sci. PhD students who are in year 7-8 and are not being pushed out the door, so this model clearly exists in other disciplines…

Let's not look to humanities for advice. If you want to talk about a field with miserable people and a pyramid scheme operation, humanities puts biomed to shame.

• GMP says:

Word.

• Dr. O says:

I don't know what world you live in, but I know very few people who can live a full life as a grad student. Hubby and I won't be able to settle down until I have a permanent job. Our life is on hold until I transfer into a tenure-track position.

And, while it's true that a continuum of critical thinking skills occurs over one's entire career, I'd argue that more time is needed during a postdoc to prepare for the initiation of one's own research program. I never would have stuck out grad school for 7-8 years. And I don't think I was somebody who necessarily should have been weeded out. I enjoyed grad school, but 5 years was long enough.

• MJ says:

"Biomed people work in large, fractious groups involving people with huge egos stomping on the peons who do the real work. NIH grants are large, but are not large enough, and are difficult to get"

That's a big part of stress it for me. And feeling like there is ALWAYS more to do. I need two other postdocs to help me so I could actually do everything I need to do (already often working 60-80 hrs and no where near getting everything done). But there's not money for that.

As a biomed postdoc I feel confidant in getting a TT position, but not in there being enough NIH money these days for me to do what I need to do to get tenure.

And re: some comments above...if I had wanted to go to med school and get an MD I would have--but I wanted a PhD. And working alongside MDs doesn't change my feelings.

• Alex says:

Many trainees take 4-5 years to finish, but some programs fill the first 2-3 years of course work/rotations (especially ones at smaller universities, in my experience). This is beneficial, but doesn’t really provide the opportunities to real develop your scientific talents.

In the physical sciences it's common to spend year 1 taking lots of courses and working as a TA, year 2 is often half research and half courses, and you aren't fully into research until year 3. People get out in 5-7 years, and while I won't pretend that it's perfect, as many people have said the physical sciences seem to be less miserable (on average) than biomed.

Pushing back the time when scientists can get a *real* job would only select against people who want to have a life outside of science…kind of not the goal when trying to select for *happier* biomed scientists. Five years is plenty of time to develop the critical thinking skills required of a PhD, even with coursework and rotations.

Five years tends to be on the short side for physical scientists. On the other hand, physical scientists tend to do shorter postdocs, and they tend not to do "postbacs" the way that many biomed folks do. So although the PhD may be (on average) longer, the overall training path is still shorter, and as I said, we may not be perfect but we aren't (on average) as miserable as the biomed folks.

• GMP says:

@Jen: So I really think the “infrequent big paper” phenomenon is rather recent, and I blame the mice. 🙂 Whaddya think?

DOWN WITH THE MIIIIIICEEEEEE!!! 🙂

Jen's comment above made me think of something. What is the role played by computer simulation in biomedical fields? I know, biomedical systems are extremely complex, but the complexity of the system never really precludes the possibility of its mathematical description at the appropriate level and consequently numerical simulation. I wonder what the status of simulation is in Jen's field, for instance?

Let me make a comparison. In my physical science field (falling under nanoscience), experiments can be extremely costly [materials growth (MBE and MOCVD machines are multimillion-dollar investments), cleanroom fees, liquid nitrogen and helium, miscellaneous equipment that's not typically shared...]. One of the rationales for the work that people like me do (theory/simulation at the microscopic level) is that a comprehensive simulation can drastically defray costs. Even industry invests in simulation because they know that developing specialized tools doesn't cost nearly as much as experimental trial-and-error, and saves tons of money in the long run. When you as a theorist know what you are doing and are focused on developing realistic simulations, you can save experimental colleagues lots of time and money by pointing exactly what not to do and what to do instead.

For instance, in Jen's field or other biomedical fields, are there simulations detailed enough to predict what a certain regimen would do to a mouse instead of actually having to develop a new mouse line? Do such simulations not exist or is it that the community doesn't recognize existing simulations as the full proof of any hypothesis, and only experiments on real mice will do?

• Dr. O says:

As many biological systems remain black boxes, I don't believe we're anywhere near being able to simulate experiments, especially those involving animals. In some very simple systems (a stripped down E. coli cell, for instance), it *may* be possible to design an algorithm that could handle what you suggest, but I wouldn't hold my breath. The majority of systems are way too complex and unknown to take into account all the possible variables.

Not that computational biologists aren't working toward this goal; it just will take some considerable time for this very young field to become useful for what you suggest. The best we can do currently is perform well-thought-out hypothesis-driven experiments and include as many controls as possible to make sure the results are meaningful...and then hope the findings are publication-worthy.

• GMP says:

Thanks, Dr. O.

The majority of systems are way too complex and unknown to take into account all the possible variables.

See, this is where I don't believe there is an obstacle of principle. There are lots of very complex systems that people weren't able to simulate at one point and now are. Complex liquids, polymers. All sorts of highly nonlinear systems. Meteorological simulation -- a humongous number of parameters.

The beauty of computer models is that you can put in and tweak at least as many and typically many more variables than in any conceivable experiment. But having good underlying models for a given level of complexity is key. From what you say, it seems that computational biomed is simply still in its infancy and it may be some time before models are sophisticated enough to be useful in predicting outcomes of animal study experiments. It may be a question of time or even computational power needed, but I would never a priori eliminate the humans' ability to simulate any complex system.

• Alex says:

The problem is, to an extent, the hierarchy of scales. In your list, the closest analogy to biology is probably weather. But in biology, there are many, many variables on many, many scales, so you need a model that can incorporate, say, tissue-level mechanics, cellular-level mechanics, diffusion in extracellular spaces, intracellular diffusion (possibly in different compartments, with possibly sub-diffusive behavior, and sub-diffusive transport can arise from many different models, e.g. recent work by Lindenberg at UCSD), reaction phenomena, gene expression, etc.

Alternately, you can model on just a few scales, and formulate questions for which (hopefully) the other scales "average out" in some sense. But that requires a re-working of how many biological questions are framed and investigated. And while that re-framing might be advantageous in some cases, there are plenty of cases where it might not be advantageous.

Of the situations you listed, polymers also tend to involve many scales, but the simulations I see tend to have interactions at one or a few scales, and then they look to see what emerges. Though I'm not an expert on it, so I may be missing a lot.

• GMP says:

But that requires a re-working of how many biological questions are framed and investigated. And while that re-framing might be advantageous in some cases, there are plenty of cases where it might not be advantageous.

Isn't that what we always do? Any good and reasonably sophisticated model will have well-defined limits of applicability, at which point a model from a completely different level of complexity becomes more appropriate.
In many physical sciences, multiscale and multiphysics computer models (bridging vastly different scales and coupling multiple physical phenomena, respectively) are rapidly coming into fashion as a result of vast computational power finally being available.

I'm not saying simulating biological systems is easy, but considering the potential for money saved, it's certainly worth it and I would never a priori eliminate it as impossible. I mean, you make hypotheses before you design experiments and in each experiment you control a finite set of parameters -- it's a model and a simplification, is it not? I don't see why biological systems would in principle be impossible to model computationally, with different models relevant for different specific subfields.

• Alex says:

Sure, in principle, all of this is possible. In practice, well, all I'll say is that there's a lot of crap out there, because a model with enough parameters can always be tuned to give the right results in some set of situations, but might nonetheless give zero insight of any value.

I won't publicly air all of my gripes about models in biophysics, but if you have some good students who understand what it really means to validate a massive computational model, tell them to consider doing a postdoc in biophysics. We desperately need them in this field.

And if you have some mediocre students who are looking to get a postdoc in a field that they think is trendy, do us all a favor and keep them far, far away from biophysics.

I mean, my latest proposal (fingers crossed on reviews) could be summed up as "Nobody has actually checked to see whether any of their data analysis algorithms make sense, so please give me to sort out this mess." Note that the lack of careful validation has not stopped people from going to market with new data analysis packages.

• Alex says:

The "no formula provided" thing was supposed to be dollar signs.

I'm all about models and their uses in my non-biomed biology field, but I would prefer that medical research is done on real animals. Simulations in biology can be great and very informative, but there's always some unexplainable error...often a great deal of unexplainable error due to the complexities of the systems. If this research is for a drug I will be using some day, I would say that the cost of an error (ie death of patients) is greater than the cost of using animal models.

Also, would just like to register that in my non-biomed biology field, experiments are often just as long if not longer (for the same reason: we use animals / plants and they do things slowly).

• Dr. O says:

I'll add that, in many cases, even animal models are suboptimal. In my field, there are constant discussions about how phenotypes might relate to real disease, when the model is so highly controlled and/or different from the human host. It doesn't stop us from doing the experiments, though, and the reasons presented here won't stop us from (eventually) using computational simulations to gain information.

I think the point to take away, in the context of the current discussion, is the vast amount of work that it takes to publish/get grant money for anything in a biological scenario where human health is concerned. When applying for NIH money, so much rides on whether or not you can prove your science is related to health, or has a chance of improving health. There's just no good way to predict this, creating a lot of pressure for biomed scientists to get their ducks in a row - for grants, journal articles, etc... The balance between publishing an entire medically-important story in one big glamor mag publication versus more incremental but solid publications further complicates the issue. And since NIH answers to Congress, who answers to a not-always scientifically-literate electorate, you never know when your science will be considered *important* enough for HHS monies. Not to be overly dramatic, but the fact that human lives are involved automatically increases the stakes, as well as the difficulties.

• drugmonkey says:

Where do you get your parameters, GMP, and which verification methods do you employ to tell you that what you have modeled has relevance? Making a more durable running shoe has a well defined outcome measure compared to fixing depression.

• GMP says:

Well, in my field, there are partial differential equations that we are fairly confident describe certain classes of systems on different spatial and temporal scales. Parameters needed for solving them on specific systems come from so-called first principles calculations as well as from experiments. Ultimate verification of any simulation tools is always experiment, but once you have extensively tested your numerical model on many, many existing experiments (i.e. made sure they reproduce what people have already seen experimentally), a good numerical model should have the ability to predict results of new experiments within a well-understood range of validity.

Of course we would write completely different equations for shoe elasticity versus wear and tear, and for what regulates serotonin or dopamine in the brain [I remember hearing a nice talk, a long while back, on difference (not differential) equations that described the mood cycles of a manic depressive versus normal mood oscillations in terms of neurotransmitter concentrations and feedbacks between them.]

To answer your question: if you can quantify a phenomenon in experiment, you should be able to come up with a mathematical/numerical model (at least in principle) for said phenomenon. If fixing depression is defined as something like "getting this neurotransmitter's concentration there within this range" then that can (again, in principle) be simulated. No one is saying that simulation should or could completely replace experiment (it never does, in most of science), but well-designed simulations can speed things up and reduce costs by reducing the number of blind alleys one goes down.

• jen says:

Modeling is Good and Important and can save both time and money - this is a really valid point. In my field there are people who do modeling on a cellular scale (actually, there is One Person). And this can and does inform experiments, and vice versa.

But, as others have indicated, we just know far too little to employ modeling on a "whole-animal physiological scale." People have tried - famously, Arthur Guyton made a diagram of everything that effects blood pressure: http://1.bp.blogspot.com/_w1oNs1aOS1s/TK6UCrs7V0I/AAAAAAAADKw/y1ZKEY0phQY/s1600/guyton.jpg

But we STILL don't understand many many things about blood pressure regulation, and people are often finding new proteins that impinge on or fine-tune these pathways in really really important ways. And literally the Only Possible Way to really evaluate these new ideas (and the only thing that will get them published) is to look at it in a mouse.

• GMP, from what I have seen in my subfield, most simulations in bio also require a lot of experimental validation. I have a friend in a physics-ish field and she spent a lot of time designing and running simulations, then picked the most optimal conditions to test. I think a similar experiment in bio would require more experimental validation of every condition that was simulated, but maybe that has to do with the black box nature of a lot of biology that Dr. O mentioned. I'm also in an area that looks mostly at individual cells rather than tissue groups or a whole multicellular organism. I imagine that modeling say, a likely response to a drug is exponentially more difficult than modeling a mitotic spindle.

• question mark says:

I believepostdoc salaries in biomed are and always have been across the board significantly lower than for postdocs in physical sciences. For example, when I started as a physics postdoc (in a government lab) on a NRC fellowship several years ago, that same year the starting salary for an NIH postdoc was about \$25K lower. In addition, biomed has a history of requiring lengthier postdoc stints as the "norm" ...although other STEM fields are now also lengthening average postdoc stints too, essentially following the path of NIH...but anyway, I think that the relatively much lower salary combined with lengthier postdoc stints in biomed compared to the physical sciences - tend to make for more unhappy biomed postdocs.

I don't know the reason why biomed postdoc salaries have always been so low, or why physical science postdoc salaries have been higher. that's an institutional thing, I suppose.

• TAC says:

I (biomed) and another student (ecology) were discussing the striking cultural differences between our departments recently, of which I was somewhat ignorant of before. (Some background : discussion started about trying to involve students in governance) We typically have a very difficult time getting any participation out of students, except for surveys where you can complain about how terrible things are. On the other hand, their department has a very easy time getting students involved, where it would be almost strange to not be highly active in your department.

One idea we discussed was the differing funding situations. Many of them have funding that is not tied to their advisor; they are expected to apply for grants, do internships, etc. which I think gives a student more flexibility and independence. Yes, they are constantly worried about not having enough money. But to work your hands to the bone over a project and to be slamming Xanax seems like an aberrant thing for them to do.

For us, it is completely the opposite. There are very few students with independent funding but the funding situation is bountiful (relatively) and structured more like a fiefdom. The attitude is more 'you should show your gratitude to me since I am permitting you to do research'. Your time to graduation is very vague and expectations vary wildly (but, of course, you too can avoid this if you work harder than them, winkwink). There is little exposure to other career options besides the post-doc, and the support is weak for those simultaneously working and attempting to supplement their career by teaching or seeking other alternative experiences (great idea! But not on my time...).

I am reasonably content; I know how lucky I am compared to everyone else.

• a says:

interesting - our biology department ranges from ecology/evolution through to biomed. I'm an ecology student.

The ecology/evolution students are by far and away the most involved with departmental activities from our grad student symposium to new student mentorship to friday evening beers.

We often see the physiologists at departmental events but never ever the cell-molecular and biomed folk (but maybe they just don't like the rest of us!)

Word.

• Bagelsan says:

I prefer working towards a Ph.D. to getting an M.D. (I think...) but I'll admit it stings a little when I am chatting with a medical fellow in the lab -- who just got back from a week in Hawaii! -- and he says "yeah, I did some moonlighting at the hospital this weekend. That is what paid for the trip." He worked for 1 weekend and that covered the cost of a week in Hawaii with his wife. Gaaaah.

He is very underpaid compared to what he could be making, if he didn't even bother with research, but he still pulls down in a weekend what I get in a month! It's a little rough to hear.

('Snot the reason for my various miseries, though; I just wanted to whine about it. :D)

• HittingTheWall says:

I agree with most of the comments above and would like to add that another reason for disgruntlement among Biomed postdocs might be the realization that after long graduate school and long postdoc stints, often halfway across the world, and crazy workweeks comes the realization that there are just too many good people left at the end of the rat race and that these people far outnumber the available faculty positions. At which point you realize that you have been pouring your whole heart and soul into attaining something that suddenly appears to be as far away and unattainable as when you were still in undergrad. And that in order to get a job outside of academia, it would have been smarter to leave earlier and work towards that from the start (maybe after the PhD but before the postdoc)...
I think one of the differences with other STEM fields is that people in other fields are far more useful in other environments and workplaces than a Biomed postdoc who knows how to rock the pipette. It's just hard to be forced to change your life's plan when you've been working toward it for so long with such dedication.

• Professor Z says:

I have heard similar thoughts expressed by biomed friends and colleagues (I am in the physical sciences) and I feel great sympathy for these people who have great passion and dedication for their work and then, through no fault of their own, 'hit the wall'. BUT I do not think that the biomed people should dominate discussion and debate, on the blogosphere and elsewhere, about whether someone should do a PhD or postdoc or pursue an academic career. (I would say the same thing for the humanities, too). This is potentially damaging (and quite unfair) to those disciplines that treat their personnel in a humane way and in which there are good career opportunities. I am not saying everything is perfect in the physical sciences, but there are many good things, including career options for doctoral students. Yet, many times when a physical scientist writes about postdocs as if this is a desirable option, the biomed people scream about how the person writing that must be blind, corrupt, delusional, and probably has a dozen postdocs chained to a bench somewhere. All doctoral students should go into their chosen field well-informed about the career options (or lack thereof), but we all need to recognize that in some cases, doing a PhD and postdoc is a very smart and likely rewarding choice.

This is a good point, and reminds me of those terrible (and frequently monthly) articles about "How to improve higher education" which are invariably written by an old philosophy professor and could only work on a small liberal arts college program...

• drugmonkey says:

This occurs when a biomed scientist (who managed to land a prof spot) suggests in any way that postdoc'ing is anything outside of the Rings of Hell too, my friend.

• Alex says:

I think one of the differences with other STEM fields is that people in other fields are far more useful in other environments and workplaces than a Biomed postdoc who knows how to rock the pipette. It’s just hard to be forced to change your life’s plan when you’ve been working toward it for so long with such dedication.

What, you think a theoretical physicist has an easier time in industry than a biomedical researcher?

Sure, we can go to Wall Street, or go do some sort of simulation or software work in industry, but there's a learning curve for us, just like I'm sure there's a learning curve for somebody who wants to go from a wet lab to, I dunno, a medical device company or something. They're both big changes.

The differences seem to be more about culture and mindset than about transferability of skills.

• MDACC student says: