Hindsight 20/20: What I really learned in graduate school

I am about to complete my ~13th year as a laboratory research scientist. This encompasses my undergraduate research, Master’s and Doctoral degrees, and two postdoc stints. Along the way, I learned a lot about the inner workings of science and pass on here some of lessons that were important to me and offer advice to those of you in the midst of your journey. If you have things to add, please leave them in the comments section below. These are derived solely from my experiences, yours may differ, and we want to hear about them!

Go to the profile of Paul Carini
Feb 02, 2017
4
1
Upvote 4 Comment
  1. During graduate school, you will grow and mature as a person as much as you will as a scientist.
  2. Learn the best way to communicate clearly and concisely with your advisor. Each advisor is different. The earlier you learn how to do this, the better off you’ll be.
  3. Your advisor is almost always busy, even when (especially when?) they are at their computer.
  4. Your advisor can’t recall all the pertinent details, protocols, results, possible directions and literature on all projects going on in the lab, let alone yours at any given moment (see #3). The most productive way to deal with this is to construct a succinct presentation for you and your advisor that does the following: i) gives a very brief background (highlighting the goals of the project or the hypothesis you are testing); ii) lists pertinent results in a logical linear order, and iii) highlights areas where you are stuck or need input. Invite other people to the meeting that might offer insight (other committee members, grad students, postdocs, etc.).
  5. Your advisor is not ‘out to get you.’ Whether you see it or not, you both have the same goals.
  6. If you are in a toxic relationship: i) ask for help from someone outside of the situation; if that goes nowhere: ii) get out of the situation. This is important in both your professional life, as well as in your personal life. Your future self will thank you.
  7. Learn how to identify other graduate students or postdocs that tend to ‘adopt’ your ideas/thoughts/words and make them their own. Sometimes this is unintentional, but there really are folks that use this as a strategy to maintain competitiveness. Be careful what you share with these types of individuals.
  8. Go to conferences. They are fun and a great way to hear and talk about cool science.
  9. Have side projects in the lab - lots of them - but not so many that nothing progresses with your main projects. Many of these side projects will slip through the cracks and go nowhere. But, a few gems will develop into papers or dissertations. My entire dissertation was a series of ‘side projects’ that derailed the ‘planned projects’ (the results of which are published here, here, here and here).
  10. Learn how to distinguish between unusual or unexpected results that are worth pursuing and those that are not. Usually, this comes down to the question “how would I test that?”
  11. Trust your crazy ideas.
  12. It’s normal to screw up. In fact, it’s expected and encouraged.
  13. It will take more time to repeat a failed experiment than to do the appropriate tests and preliminary experiments before setting up the 'big experiment.’ Put another way: there are no shortcuts.
  14. Science is expensive; how much do you want to learn? That’s not to say that money should be wasted, but don’t disregard an important line of inquiry because you are worried it will be expensive. It’s likely you have a different definition of ‘expensive’ than your advisor.
  15. Publish your work early and often. This is analogous to the adage: "If a tree falls in a forest and no one is around to hear it, does it make a sound?" If experiments are conducted in a laboratory and no one publishes them, was any knowledge gained?
  16. Publishing papers takes an enormous amount of time and effort. A huge lesson for me was that it takes several months or up to a year after the paper is submitted to see it published in a journal.
  17. You get amazing manuscript feedback when you present your advisor and co-authors with a first draft that represents your absolute best work. In my personal experience, when a colleague is given a draft that looks like a real paper – that is, it is written clearly and concisely, fully referenced, contains publication quality figures and tables (with captions) and has all supplemental files attached – they will put your paper near the top of their ‘to do’ list. It is also very helpful to point out specific areas for co-authors to address/help with comments. That said, if you get stuck and need input before the manuscript is finished, ask for help (see #4, above).
  18. Read Day’s How to Write and Publish a Scientific Paper and Strunk and White’s Elements of Style before you start writing or when you get stuck.
  19. Try to avoid writing manuscripts, and especially dissertations, in Microsoft Word. Only export near complete drafts to word so collaborators can comment. Personally, I compose all manuscripts in Scrivener and highly recommend it, but there are several other useful programs designed for writing manuscripts or other large documents.
  20. Obtain a reference manager. Keep the metadata for each imported manuscript up to date from day one. I curate the metadata the moment a paper is imported. Similarly, do not blindly trust these programs to format your references 100% correctly when compiling a manuscript. Always double check the references. I commonly find 1-2 mistakes. I prefer Papers 3 after trying several reference managers.
  21. Buy (or otherwise secure a copy) of Adobe Illustrator and learn how to use it.
  22. Learn how to analyze data in R. The learning curve is worth it.
  23. Learn how to cope with rejection and criticism. This is very hard to get acclimated to, but developing rejection coping methods is essential. Go for a run, watch Friends re-runs, play some Slayer cranked to 11…. whatever works for you.
  24. Celebrate victories. Because rejection is so common in science, it’s important to celebrate each paper accepted, grant funded and successful experimental result.
  25. I was much happier as a graduate student once I learned to ride the waves of #23 and #24. That is - don’t get too low with bad news and don’t get too high with good news. Try to maintain an even keel.
  26. Don’t procrastinate. Easier said than done, I know.
  27. By the end of your graduate career, you should be able to think and plan for things 6 – 12 months into the future. For example, it takes at least 6 months to sift through and vet potential postdoc positions.
  28. Very few people care what your grades are in graduate school. There are many advocates of doing the least amount of work in a course to achieve the minimum required grade. Focus the extra energy on your research.
  29. You do not need to be a ‘straight A’ student at any level of your education to be a successful graduate student or scientist. It is more important to be creative, observant and to work hard.
  30. Take care of your health. Regularly exercise and eat well.
  31. Drink enough, but not too much, coffee and beer.
  32. Science can be stressful, but it’s important to keep perspective. Science is only one small aspect of who you are as a person and as an individual.


Poster image credit: Nimish Gogri

Go to the profile of Paul Carini

Paul Carini

Assistant Professor, University of Arizona

Microbial cultivation and functional genomics.

1 Comments

Go to the profile of Michael Chao
Michael Chao 10 months ago

Thanks for the post; it really brings me back to those good (and stressful) old days. One thing that I would tell any starting grad student is that finding the right lab is more important than just finding the right 'question'. You can learn to be a good scientist in many different labs, and being too caught up on a specific topic area might limit your ability to find a better fit in another lab (in terms of personality cohesion, level of support from other members, ease of learning, etc). Opportunities abound to pick up knowledge about lots of different areas (rotations, seminars, collaborations, journal clubs), but finding a place that feels right and will work for you for 3-6 years is something you should prioritize.

Also, wanted to point out that several points are also relevant for navigating scientific publishing too:

16. Publishing a work is indeed a long process, as your peers are both your friends but also top scientists who are tasked with ensuring the conclusions of the published work are sufficiently supported. This often requires a long time devoted to additional (and sometimes unexpected) experiments and revisions. As such, point 13 is also good to think about before going in--given the choice between an indirect but easy experiment and those that can more directly address the question at hand (but may require a lot more work), most scientists reading (and reviewing) a paper will likely want the latter (even it's more difficult); it might be easier to nail the point with the most informative experiment upfront.

4. Constructing a paper with a straightforward narrative that includes clear rationale enables both editors and referees to understand the work and identify salient points and critical points requiring additional work; in particular, what might be second nature to the researcher could be presented in complex leaps of thought that are hard to follow for others (even other experts), which can lead to more confusion about the overall conclusions. Having others read your drafts is a good way to deal with this, but point 9 is also relevant--side projects and tangential findings can be fun and important, but think hard about whether they fit the goals of the main paper and should be included, as they can distract from the main message and sow confusion down the line.