Your project will probably fail.
While discussing projects with my mentor some time ago, he shocked me by disclosing that typically the most arduous and hi-risk projects are given to the most inexperienced and least likely to succeed, i.e. graduate students. Why? Because the alternative is to give those projects to post-docs who have a lot more riding on their success in the lab. Apparently, as Grad-students we can ‘afford’ to fail…
Failure and risk are unavoidable in science and even more so in gradschool. Now, there are several ways your project can shit-the-bed.
1) The hypothesis can be wrong
2) The discovery can be uninteresting
3) The required experiments can be technically impossible
4) you can get ‘scooped’
There are some things you can do to avoid having to dump years of hard work and start over, but I think the most important thing is to have a back-up. It is critical to have a second project simmering on the back-burner in case your ‘main’ project tanks. To make your life easier the projects should be related, but be careful that the death of one doesn’t pull the other to the grave with it.
Avoiding failure requires forethought, and lots of it. For example, it is possible to devise a project such that the research is interesting and publishable whether your hypothesis is wrong or right. For example, your data suggests that some interaction of proteins happens in a cell, a good project will be able to say ‘Hey, I found a novel interaction that does X’. But if the data goes the other way it could say ‘Hey, X unexpectedly happens without the interaction you would expect’. It’s not always easy or possible to devise such a scenario, but it should always be looked for because it will greatly increase the security of your project.
Likewise, making sure your discovery is interesting requires discipline and forethought throughout the research process. Finding an interesting molecular interaction inside of cells is great, but how great is it if in the end you discover that the result of that interaction is redundant and only marginally different from a process that has been shown ten times over? To avoid this kind of situation I think it is best to either start with a solid and relevant (preferably clinically relevant) finding, and/or jump ahead to the punch line. Figure out the clinical or biological significance (that people will care about) about your molecular interaction before teasing out all the details in between. Real science, although presented as such, is seldom linear.
Technical challenges abound in science. If it were easy it would have already been done, wouldn’t it? Going into a project that requires only techniques that are well established in your lab will save you lots of agony, frustration, and potential failure. However, novelty is the drug of choice among scientists, so employing a novel technique will boost the interest and excitement of your project. The key, I think, is to know what you are getting into. Think about the technical challenges you are creating when planning your project, try to limit them to a reasonable number, like one, and have an alternate method as a back-up.
Getting scooped is the great and perpetual fear of all researchers. One day you’re knee deep in the throes of research, the next you see that someone published exactly what you were working on, and did it better… The only sure fire way to avoid it is to alienate yourself from your friends and family, forget what ‘vacation’ and ‘holiday’ mean, and work your butt off 24/7. However, while you may have little control over whether or not you get scooped, being aware of how competitive your field is will help you know how to pace yourself. Highly competitive fields may require you to run closer to that line of burn-out, and you need to be prepared for that.
Lastly I’ll say that if your project bombs-out, please, don’t panic. It happens to all of us all the time. Take a day off to eat some ice cream, punch a bag, or question your sanity in deciding to go to gradschool, then come back to the lab and find your new quest.