The road to a science Ph.D.

By Jonathan M. Gitlin

Monday, September 19, 2005

Starting this week, I'm going to take a break from just reporting on interesting new developments in the world of science to take a look at the institution itself. I can't count the number of times I've had young children rush up to me and ask, breathlessly and with hope in their eyes; "Hey Mr. Science Man, how do we make it big and win the Nobel Prize?" Actually, I could count the times if it had ever happened, but I'm going to keep pretending it's an everyday occurrence and therefore the basis of this piece.

The road to scientific glory and the plaudits it brings is a long one. I'm going to exempt the earliest stages, from reading books like Charlie Brown's Big Book of Questions and Answers (my own start in science) to activities like school science fairs and work placements in laboratories. I will mention, as a benefit to younger readers or parents who might want their children to get into research, that fainting at the first sight of a dissected mouse and then never coming back to the lab is a bad start to a summer internship.

From here we venture on to university. At this point, transatlantic differences begin to leap out, and I must ask my US readers to bear with me should I err. Once at university, you begin to learn that most of the things you've been taught up until now are wrong, or maybe they're right, but not for the reasons you've been told. You'll also learn that staying up late and drinking too much is fun, but getting up early, wearing a tuxedo and drinking too much is even better.

Depending where you're studying, you might even get your hands dirty on occasion with an experiment or two. However, budget cutbacks and increasing class sizes are a limiting factor for gaining hands-on experience. In recent years in my own field, computer-assisted learning has been the big thing, and instead of a real piece of guinea pig ileum twitching away in an organ bath, it'll be a virtual one. Better, perhaps, for the fluffy rodent, but much less chance for 'accidentally' poisoning your labmate.

In order to combat the allegations that the universities were turning out a generation of graduates without a clue when it comes to working in a lab, most science programs offer an extra-mural year, wherein the plucky young undergraduate heads off to work in a lab. Armed with bright notions about just how smart they are and how they'll be given their own lab, the institutes and companies that receive these precocious neophytes are just as pleased, knowing that they've got a source of cheap labor for the next twelve months, and point them in the direction of the washing up.

Despite the fact that one often gets saddled with the crap jobs, the extra-mural year is a vital step in the training of a scientist. Away from the classroom, essential basics such as how to make a serial dilution, what to do when you spill radioactive things on your jeans and how dry ice and eppendorfs make good booby traps are learnt through experience and repetition. Assuming you get a chance to work on more than one aspect of a project, it's also a good chance to start finding out which bits of science appeal and which don't. This is important, as there's a long way to go before you'll be done.

So you head back to Your U. to finish up and gain that precious B.Sc., delighting the family and surprising your doubting school teachers. You've learnt some practical experience working in a lab, you know one end of the forceps from the other and you're not running off to become an investment banker. So, next stop: Karolinska Institute, Sweden? Well, not quite. Having a bachelors degree in science is essential, but only having one can be detrimental, depending upon one's desired endpoint. If you're happy to work in a lab, have a certain amount of job security, want a pension and benefits and don't feel like spending the rest of your life writing grants, then now is a good time to apply for jobs. I am not going to say anything bad about technicians—I've worked with some marvelous techs, and they all got paid more than I did. But if you want to one day run your own lab, or work with much autonomy, then you're not done yet. Not by a long shot. Off you go to graduate school!

Sometimes people complete a Masters degree, but this isn't always necessary, and so we find ourselves as PhD students. Just how you find that PhD course (or program) isn't always easy, but there are a few tried and tested methods. Positions are advertised in the back of Nature, Science, New Scientist and online. Universities let each other know about positions, and finally the networking that you did in your extra-mural year might just come in handy too. Choosing a PhD course wisely is vital. The institute name will follow you around on your CV, so prestige helps. The work environment is important too. You will be spending at least three, and maybe as many as seven years there, so pick one that's not going to have you spending all your time in a damp and draughty basement. Your mentor/advisor/supervisor/principle investigator (PI)/boss is the last and most important step, and one I cannot emphasize enough. This person will likely become the template for you in future life as a scientist, but if they're a crazed megalomaniac the next few years could be hell on earth.

So, you've found a PhD course. It's here that cultural differences begin to become greater, as US graduate students discover another 2+ years of classes and exams, but in the UK, Australia and other parts of the world, day one involves you, a fresh lab book, and, if you're particularly unlucky, being locked in the supply closet. Life is much like it was as an extramural student, with a few differences. When a new undergrad turns up in the summer, you get to boss them around and lock them in the supply closet. There's more personal responsibility, they pay you (slightly) more, and every so often the department head will ask you something in a lab meeting, giving you the chance to shine or embarrass yourself. Otherwise things haven't changed much. Techs still look down on you, postdocs look down on you and the faculty do too.

Now, a strange thing starts to happen. If you've got this far, it's safe to say you're probably feeling a little proud. You're very bright, or at least you think you are, and walk around campus pretending to be Ice Man. Pretty soon however, it begins to sink in that you're not such a big fish after all, and the pond can look more like a lake. Not only are there other people here that are as smart as you, some of them are actually smarter and others that work twice as hard. This shock can take some time to overcome, and often infuses a little humility into ones character. Other times it might contribute to becoming the aforementioned megalomaniac.

I'm not sure what you do all day if your PhD is in one of the humanities, or something like maths, but in life science, it means experiments, but this can mean many things. You could work on a clinical project. This is good if you don't like working with animals, but has a few other drawbacks. Firstly, you're at the mercy of the clinicians who collect the samples. There's sometimes an innate resentment between scientists and medics. They get to call themselves Dr without an advanced degree, and yet they're the ones who always get made department head. You might be strictly cell culture based, which means you get to spend your days sweating in a cell culture room playing with plastic flasks and plates full of red/pink liquid. (Always remember, spray OR flame, never both!) Also, now's the time to switch to nitrile gloves if you might develop a latex allergy. You might even do such cool work that you end up in a BSL (biosafety level) room. I don't know if they let PhD students into BSL4 rooms. I hope not.

You'll probably have to work on animals at some point. Mice and rats are most common. Some labs use bigger animals, but cost and cuteness tends to make that more and more uncommon. Other projects could see you working in the field, chasing antelope or sleeping with meerkats. It probably doesn't bear saying, seeing as you're reading this, but proficiency with technology and computing devices is a serious advantage. The glory that is Pubmed revolutionized life as we knew it when it came to keeping abreast with the literature. If you took a series of time lapse photos of the fellows room (in the US a common room) I spent my time in, you'd see two old and unreliable PCs running Windows 98 and shared by 12 of us become a laptop on almost every desk, but expect to see a lot of two-finger typists.

Knowing your way around Powerpoint is vital. It is not uncommon to be accosted by a professor shortly before they're due to fly off to an important meeting, demanding or pleading for someone to fix their $%^ing laptop. Less common, but out there, are the faculty members who know a thing or two about technology. If possible, these are people you want as your boss. The lab will have up to date computers and technology, and you might get to play with some cool toys. A certain promiscuity with platforms can be propitious. It's great being a mac person in a mac lab, but being the sole maclot amongst a sea of Dells can have its disadvantages. Likewise, the lone start bar in a herd of apple menus can be a lonely place to be. Expect to frequently come across antiquated computers connected to important pieces of equipment that are in all probability aren't even networked. You'll think about trying to upgrade the hardware and OS. The thought of important data being captured under OS 8 or Windows 95 is a scary one, but the program you need will have ceased development, or it'll cost extortionate amounts for a seat license, or maybe everyone else made do with it and they want you to suffer too. Some buggers even spend all their time in front of a screen. The rise in bioinformatics and the sheer mass of data that comes off a microarray means that database miners will always be busy at the virtual coalface.

The volume of work one is confronted with also seems up for debate. Common wisdom is that you'll spend all your time in the lab. Evenings, nights, weekends. There's another, more controversial school of thought that says this doesn't have to be true. A lot has to do with the type of experiment you conduct. Some might take four or five hours over two days to yield results. Others might involve spending 10 hours at the bench in one sitting. As you go through your orientation at the beginning, the importance of planning will be come up. This really is crucial. With successful planning, it is perfectly possible to have a productive week in the lab with time replete with late nights out, late mornings in to work and everyone in the pub by 3 on Friday. Important things to note: always be in early for lab meetings. Nothing draws attention from all the wrong people like being late. As I mentioned earlier, department heads are often clinicians. That means they like doing things at ohmygoditsearly o'clock. Being a first year PhD student ensures that, when you go to the pub with your department, they'll probably buy your drinks. Don't abuse it, and repay the favor as you move on up the chain. Writing down your results is vital. Knowing where you've written them even more so.

The next few years are not always plain sailing. Friends who graduated with you move on to high paying jobs whilst you remain in relative penury. Experiments don't always work out. There are assessments and committees to deal with. If you're unlucky enough to have classes, there will be homework! Although you're researching a particular niche of your own, you are also being formed in the same crucible as the scientists who precede you. They all had a tough time of it, and they'll be damned if they'll let some young upstarts off the hook easily.

I mentioned earlier the beauty that is Pubmed, and the way it's transformed the way we can access the literature. Prior to the 1990s, if you wanted to search for a reference you had to pick up a volume of Index Medicus and dig through it. Although you can now download PDFs of journal articles from the comfort of your office chair, it's still worth getting to know the dusty stacks. Journals will send out e-alerts and now some are using RSS, but not everything is available online, and midweek hangovers respond well to an hour or two in the library catching up on current journals, reading the New Scientist or Scientific American and making silent pacts with various deities regarding the wisdom, or lack thereof, of mixing your drinks.

It's about now that conferences start to take on importance. These come in various shapes and flavors, from small local meetings to the largest international ones. Conferences serve several functions. You get to show your work to your peers, some of whom will try and make you cry with hard questions you can't answer. Sometimes, you get to present something that's in direct contradiction to the person who was up before you. Showing your work involves either a poster presentation or giving a 10-minute talk. Opinions are divided on their merits. Posters can be less nerve wracking, but you do have to stand by them for several hours. You might be constantly swarmed, or you might get ignored and resort to busking to draw attention to your breakthrough discoveries.

Giving a talk can be somewhat terrifying, especially if you lack confidence in your work or your abilities. It is over much faster though, and has more opportunity for guts and glory. The thing is, giving presentations is part of the territory, so the more you do now, the easier you will cope in future. In my lab, our boss looked down on posters as an easy way out, and therefore each abstract we submitted was marked (T). By number 9 I had found most of my fear at being in front of my peers and having my work ripped to shreds had evaporated.

But conferences are more than just about 10 minutes of sheer terror or 3 hours of sheer boredom. First, it's time off from the lab. It's a chance to meet up with old friends and make new ones. Networking is important, and I'm not talking about wifi. Just like in every other industry, this is how the best jobs are found - word of mouth. You'll also get a chance to meet the opposite sex, and odds are they're going to be dressed up and showing off like peacocks. Conferences are also an opportunity to meet other people working on similar problems to you. Finding out that everyone has the same problems with a particular assay, or developing a collaboration with a lab that can complement your work are ways in which a boring meeting can become a great success.

OK, you've beavered away in the lab for a few years now. You've transformed from a swaggering, cocky upstart to an overworked, underpaid cynic. You may have endured a crisis of confidence, bouts of depression or maybe even turned vegan. Experiments work now. Data begins to build up and you start putting together a picture of your thesis. Writing up affects people in different ways. Some do it quickly, over a matter of several weeks. Others find it a much more drawn out process. This author obviously enjoys the act of writing, or you'd not be reading this, and so I found it true for my PhD. Sitting down and surveying the body of work you've produced over the preceding years lets you appreciate the scope of your work, and you can show off and use long words. Different mentors will encourage their young charges to approach the writing in different orders.

A thesis will consist of the following: A general introduction, methods, results chapters and a discussion. Word and Endnote will take care of the references for you. I'd start with the methods. This describes what you've been doing everyday in the lab, and ought to be the easiest part. It is safe to say that if you struggle here, you're going to have bigger problems further down the line. You can even start to write this as you work your way through grad school. The general introduction is where you get to read over all the important papers, and many more too, in the field, in order to put your work in context. The thing is, this will be read by your examiners, probably quite carefully, so for god's sake make sure that the papers you quote actually say what you think they say. Next up come the results chapters. Each of these should tell an individual part of the story, and each one might make a paper in and of itself. Be consistent with your graphs and figures. It's here that those IT skills will begin to shine again, and you can fill your thesis with schematics and diagrams while you laugh at your colleagues, brows deeply furrowed, hunting and pecking away at their keyboards. The discussion will be the last part you write, wherein you explain the greater ramifications of your work, and make wild statements about how you think it will transform the world, without having to actually back it up too much. I had great fun writing mine, and was thoroughly looking forward to discussing what I saw as brilliant thinking and writing. Boy was I in for a surprise.

Next week, we look at the ordeal that is getting your PhD, and the excitement to follow.

In other news:

• The Guardian takes a look at bad science and bad science reporting. One of the things that poor science education results in a public who don't have the ability to tell when a journalist is misunderstanding or misrepresenting the truth. This is more and more important given the concerted political attempt to muddy the waters in certain fields.
• Hurricane Katrina has had serious consequences* for researchers based in New Orleans. The flooding and extended loss of power meant many -80ûC freezers thawed, and all the animal research facilities at LSU were lost. In 2001, Tropical Storm Allison caused flooding that drowned all of the research animals at the University of Texas Medical School in Houston.
• Has Titan's sea finally been found? Cassini is now sending back images of what ap-pears to be a distinct shoreline, with a dark region believed to be a sea of liquid methane. In other astronomic news, a gamma ray burst detected at the beginning of the month originated 13 billion light years away. That must have been a hell of an explosion.
• Is it too late to actually do anything about climate change? Yes, according to a report in the Independent, a leading British newspaper. According to the researchers interviewed, the melting of the polar ice cap in the Arctic Sea has probably passed a tipping point. As long as the world's largest polluter continues to deny or downplay man's role in this slow walking disaster, and it's most populous nations strive to attain the same kind of industrialized behavior, then I'd probably agree, it is too late.

* Subscription required