Living the Life - Perspective of a Grad Student

One tremendous perk of being a graduate student is that you have time. Time to learn how to do science and savor the experience. Time to find your passion and refine your interests. And most importantly, time to learn what it takes to lead a successful lab.

Below are my notes from grad school on the different roles in a lab and what each one means. I have purposefully left these in their original form to remind myself of the vantage from a PhD student.

Graduate Student

"Being a graduate student is the best." - Wallace Marshall

You have no responsibilities other than to follow your heart and find the type of science you are most passionate about. You utilize the university setting to take fun classes (drawing, storytelling, improv) and marvel at the fact that you are now trusted to teach the next generation (who, after all, are only a few years younger than you are). You collaborate with and are constantly sheltered by your advisor. With a 1-minute commute to work, and having just received the biggest pay raise of your life (from paying for tuition to being paid for your services), your life is unbelievably amazing!

Postdoc

"Being a postdoc is the best." - Wallace Marshall

Postdocs know everything. Seriously. Tell any postdoc about your work, and they will suggest several relevant papers. They ask deep questions during group meetings. They know most aspects of their work better then their advisor. Clearly, at some point between the end of graduate school and the beginning of their postdoc, every scientist becomes orders of magnitude more knowledgeable. It is likely a gradual (and hence unnoticed) transition, but it definitely happens.

But postdocs are also cautious. Most are hesitant about committing themselves to academia, since there are many more outstanding candidates than there are positions. Some go into industry, while others stay in school forever. In either case, it is incredibly exciting to see someone you met as a postdoc move on to bigger and better things!

Professor

"Being a professor is the best." - Wallace Marshall

Professors are incredible. They know everything and everyone. They are the top of the food chain, leading the charge on important problems. Yet when asked what their favorite part of their job is, every professor will tell you it is seeing their students grow and succeed as scientists in their own right.

Professors have the honor of teaching as well as convincing other scientists that their work is revolutionary. They have great control of their lives, and they take a year-long break every 7-ish years to reinvigorate themselves, reflect on what their lab has accomplished, and reorient their goals to better match their passions. They have the privilege to lead a tight-knit group of incredibly talented individuals to tackle problems that no one in the world knows the answer to. And that is, unquestionably, the coolest job in the world!

Ten Traits of a Successful Graduate Student

Graduate school follows a "choose your own adventure" format, and there are many ways to play the game. Yet the best students generally follow similar strategies.

Love what you are doing: In a perfect world, just thinking about your job should elicit a contented sigh. If you do not absolutely love your current project, ask your advisor for another one. Do not be afraid to reject a project that your advisor offers! There are a huge range of problems to explore, and there is undoubtedly something that both you and your advisor find interesting. Keep searching until you find a question that gets your blood pumping, and then go out and conquer it!
Interact with your lab: Be nice not only to your advisor, but to your fellow graduate students, the lab manager, and every staff person that you will ever meet on campus. You will interact with this crew for the next five years of your life. You are never too smart to be a jerk to someone. Everyone is busy, but everyone is happy to help someone who is positive and polite.
Work smart and work hard: You do not have to reinvent the wheel. Read a manual on your favorite programming language. Learn about the cool new text editor. This also applies to research – read the literature to understand the context and implications of your work.
Take the initiative: As an undergrad, you could be a code-monkey, merely following instructions without knowing what the next step should be. In graduate school, you have to figure out the next steps and the applications of your theory. If you find an interesting paper that could motivate your next project, jump on it!
Be thorough: When told to do ten things, do not let any of them fall through the cracks. Apply lessons learned in all relevant contexts from then on.
Ask questions: At all times. Whether or not you get a satisfactory answer, write it down. After five years, these notes may inspire some new insight, or at least be good for a laugh.
Learn new skills: This is part of your job – how cool is that! Abandon your old prejudices (such as, "I have always hated English class and therefore dislike writing") and embrace every opportunity to learn more math, figure out how to improve your scientific illustrations, or upgrade your website to include the top ten traits of a successful graduate student!
Talk to other people: Do not isolate yourself within your lab or building. Interact with other graduate students on both a professional and personal level. If you are stuck trying to solve a complex problem, chances are that a smart math graduate student already knows the answer. Science is a community – do not go through it alone.
Stay organized and efficient: This might seem easy when you start and are only working on a single project, but think about what will happen in five years. Every few months, spend some time reorganizing. Automate as much as possible to maximize the time you spend being creative rather than in tiresome drudgery.
Be optimistic: This year will be better than last year, but not as good as the next year! Believe that your experiments will work, that your ideas will pan out, and that your paper will make a splash. Any feedback you get – even the seemingly cutthroat criticism from Reviewer #3 – will make your work stronger, provided you do not dig your heels in and refuse to change a thing. Always smile, and everything else will work itself out.

Personal Statement

With a few innocuous words, my life radically changed in middle school: “In this class, it is just you and the computer. No teacher. No instructions. We have a series of exercises, and your goal is to get through as many as possible. Have fun!” As I sat down in front of a blank white screen with Mathematica written across the top, the first session was a confusion of brackets and capitalization errors. But within a week, this maelstrom turned into excitement and then to wonder.

In my mind, each new exercise seamlessly transformed into code. Yet as I realized how many ways there were to solve the same problem, I tried to envision the landscape of all possible algorithms. I could take small “steps” up to improve my code, but when I reached the top…what would I see? I imagined looking out at sprawling mountains, representing fundamentally new ways of coding, and I wanted to go exploring.

During graduate school, I took a non-traditional path, opting for a 1-year sabbatical between the 1st and 2nd years of my PhD (so I could tailor this year based on my specific field). I chose to join Wolfram Research as a software developer and work on my favorite programming language, Mathematica. Although I already had 10 years of coding experience, my skills rocketed upwards during my time at Wolfram. Co-workers would routinely send emails describing the problems they were trying to solve, and while I could readily answer any question with a paragraph of code, experts in the company would respond with compact 1-liners. These were masterworks – akin to Picasso's line art – that were not just beautiful, but provided a glimpse into those faraway peaks that I had imagined in middle school.

Fast forward to the present, where I use programming every day to understand antibody-virus interactions. Each step I take is magnified a million-fold, cutting great swaths through the space of possible models. And I am no longer exploring just one mountain: machine learning has expanded my toolkit, and the growing list of machine learning interpretability tools help elucidate actionable lessons from such methods.

My most surprising discovery during my postdoc is that new results are everywhere! With each new computational technique or dataset, I find a startling trend or a conserved property of the antibody response. This “commonplace novelty” had solidified my belief that we are now data-rich (and growing richer by the day).

There has never been a better time to be a computational biologist. Both the number of biological datasets and the number of Mathematica functions have grown exponentially in time, providing countless ways to combine them together. My work is a blend of art and science, of intuition and intellect, of mastering modern tools and developing new techniques. And I would not want it any other way.