On my Love of Science and Why I Left Graduate School

The first time that I set foot in a research lab as an 18-year-old college freshman at Clark University, I knew that I was home and that I wanted to be a research scientist. That fall, I set to work synthesizing unusual and previously undescribed magnetic compounds from mixtures of copper ions and an assortment of organic molecules. I loved watching the chemical reactions in real time, the sound of the automatic stir bars whirring against the edges of my beakers. I loved the smell of the organic molecules hanging in the air and the quietude of the lab late at night. Afterwards, over a period of weeks or months, crystals of new compounds formed at the bottom of the beakers. They were beautiful- bright oranges, vibrant yellows, chartreuse and forest green. I relished being the first and only person with a supply of these new compounds.  If the crystals reached a sufficiently large size- barely a millimeter across- I would scrape each crystal up into a vial and ship them off to collaborators a half a world away in New Zealand. There, scientists subjected my crystals to powerful x-ray radiation. The photons bounced off the electrons of each atom in the molecule, creating a unique diffraction pattern for each compound. They would then send the diffraction data back to our lab in Massachusetts where we would use it to solve the atomic-resolution structure of the compound for the first time. In parallel, I took magnetic measurements of crystals that had never been shipped to New Zealand. To do this, my professor and I cooled our compounds to nearly absolute zero, placed them in a magnetic field tens of thousands of times stronger than that of the Earth, and slowly raised the temperature to see how the electrons would align themselves. Learning the molecular structure and magnetic properties allowed us to glean insights into how subtle differences in molecular structure could lead to radically different magnetic properties. In the long term, this work builds scientific understanding of how to create superconducting magnets that could change the face of technology in ways both subtle and life changing.

After three years, my imagination captured by x-ray crystallography and swimming with new knowledge of biochemistry, I decided to change direction in my research. Full of ambition and with my eye on combined MD/PhD programs, I drove across town and joined a lab at the University of Massachusetts Medical School which was laser-focused on uncovering the mysteries of a cancer-causing protein nicknamed CtBP. I quickly learned of our complicated relationship with CtBP, a protein that is both a savior and executioner. If an embryo suffers a mutation in their CtBP genes, the pregnancy is not viable and quickly ends in miscarriage. However, transcription of CtBP is rapidly shut off after early pregnancy, never to be turned on again. Unfortunately, in a subset of cancer patients, the body erroneously ramps up CtBP production. This in turn makes the cancer cells harder to kill and allows them to break away from existing tumors, and buoyed through the blood, land in distant tissues, colonizing more organs with a smattering of growing tumors. As of now, there is no treatment for this phenomenon. We had a singular objective: understand the structure and biochemistry of CtBP to make a drug that renders it inert in cancer patients.

As before, I quickly fell in love with the day-to-day research. I inoculated E. coli bacteria with synthetic variants of CtBP genes, and then while growing trillions of bacteria in large flasks of broth that smelled vaguely of chicken soup, I used a chemical to induce the bacteria to produce unnatural amounts of CtBP. Afterwards, I broke the cells open and purified the desired protein. In some experiments, I conducted batteries of biochemical tests to understand how individual CtBP molecules bind together in the hopes that we could design a drug to interrupt these interactions. In other experiments- my favorite experiments- I worked to crystallize the purified protein in a similar manner to the way I crystallized the magnetic compounds. I took microliter droplets of protein under varied chemical conditions, used a pipette to carefully transfer the droplet to a plastic microscope cover slip, and then flipped the coverslip upside down and sealed it off from the atmosphere and waited. Over the course of days and weeks, CtBP crystallized under a small range of precise conditions. The crystals were only visible under a microscope and were uniquely beautiful, gleaming like nearly translucent diamonds cut into a bipyramidal shape. While looking through the microscope, I added miniscule amounts of experimental drugs to the droplet and watched as the crystal cracked and then annealed when the drug bound. From there, I immediately scooped up the crystal with a tiny spatula and mounted it on a specialized x-ray machine. For hours thereafter, we took hundreds of x-rays of the crystal from different angles, slowly building the diffraction pattern needed to solve the atomic structure of the protein. In the following weeks, I poured over the x-ray data, slowly building an accurate structure of CtBP. Although I only solved three structures despite setting up over 3,000 protein droplets, the work was deeply gratifying. I decided that I wanted to spend the remainder of my career studying protein structure to advance medicine, abandoned my fantasy of going to medical school, and instead applied to PhD programs in biophysics and structural biology. I was at the peak of giddy confidence as I wrapped up my work at UMass with a master’s degree and a publication, ready to throw myself doggedly into the next stage of research.

I came to the University of Pittsburgh in the fall of 2017 in search of a new lab and a new scientific problem. The beginning of a PhD program yields the opportunity to reinvent oneself- to discover new ways of thinking about research, to deepen one’s existing knowledge, and learn new techniques. It gives one opportunity to try and fail, and to try again. At this juncture, I was infatuated with academia.

I approached my work with the usual zealotry, sometimes going months while working seven days a week. I lived on ramen, caffeine, and Chinese takeout. During the first eight months of my program, students rotate through three different labs in an effort to match with the right dissertation adviser. In my case, serendipity heavily shaped my decision. During my second semester, I was forced to drop out of a required computer programming course because I developed a mental block under the strain of the high workload. Panicking slightly and searching for alternatives, I asked my first-year adviser if I could rotate in her lab. She is a computational biophysicist, and all her students become adept programmers as they simulate proteins to understand how they fold into complex 3D shapes and build new software to simulate a vast array of chemical processes. Always better at research than classwork, I thought that hands-on experience with coding would get me back on track.

I immediately became fascinated with simulating proteins. I threw myself into the research, hungrily learning about this new approach to studying protein structure, and quickly made inroads into the intro-level problems that I was initially assigned to learn the techniques. I regained the confidence that I had lost when I dropped out of the programming class. After two months, I knew that I wanted to change directions in my research and study protein structure computationally.

I relayed this insight to my adviser, who then suggested a once-in-a-lifetime opportunity. She correctly intuited that I missed experimental work at some level and came up with the idea of a co-mentorship. The plan was simple: I would split the time between simulating proteins in her lab and working with proteins hands-on in a second lab under the tutelage of a renowned experimentalist and close collaborator. Towards the end of my PhD, I would tie these separate strands of research together into one momentous capstone project.

I knew that accepting this offer was a bit of a gamble- a co-mentorship is a particularly grueling track, especially since I was still a novice at simulation techniques. Nonetheless, I could not turn the offer down. Fueled by a mixture of equal parts curiosity and blind ambition, I delved readily into this unfamiliar situation. On the experimental side of my work, I began attempting to purify the HIV protein Vpr while it interacted with the human protein hHR23A. In my computational efforts, I began simulating protein unfolding mechanisms to gain fundamental insight into how proteins spontaneously lose their structure.

However, I quickly ran into problems in both labs in the summer of 2018. On the computational side, I was having difficulty learning the mechanics of programming beyond the introductory level. On the experimental side, I quickly came into conflict with the lab manager, who refused to provide a set of pipettes, insisting that they were needed for visiting undergraduate students, even though they are a central tool in any protein lab. I compounded this conflict when I borrowed his pipettes late one Saturday night and forgot to put them back in the right place. I was angrily confronted on the following Monday, and it was immediately clear that I had committed a serious taboo and that my standing in the lab was in trouble. I was ashamed and angry at myself for weeks. I had been the lab-equivalent of an inconsiderate roommate, or so I felt at the time.

Work intensified through the early fall when I began the last class required in the core curriculum. Here, we bounced between studying thermodynamics and statistical mechanics, solving problems with pen and paper, and undertaking lengthy coding projects with little oversight. Working close to 100 hours a week, I was reaching my breaking point. I was neglecting my family, my friends, and my soon-to-be fiancée who had just moved 600 miles from Massachusetts so we could start our life together. I was not taking care of my health. I was not so much as making progress but putting out fires. Everything came to a head in late September, after I asked a question in class which the professor couldn’t answer. Clearly annoyed, he assigned me the extra homework of answering this question and presenting it to the class. When I did not have time to complete this additional piece of work, he mocked me in front of my peers and told my classmates that it was my fault that this question was going to be on the exam. In this moment, I felt the fight go out of me. 

Rather than going back to work after class, I took a long walk to the public park, and while sitting on a bench atop a peaceful hillside, decided to leave science. In that moment, I felt the sharp, unremitting anxiety of the past months die away, while in its place grew a sense calm, if not serene, depression.

I emailed my advisers, explaining my situation and decision to leave. Alarmed, they called a direct meeting. To my surprise, this was not an exit interview. Instead, they offered six weeks of paid medical leave so that I could recoup my mental health. I could drop the class and make the credits up in the spring with a different professor. They explained that the school understands that a predictable fraction of PhD students will suffer a nervous breakdown, and that there is an informal institutional flexibility in place to accommodate for this. Stunned, I accepted the offer and decided not to look for jobs.

My time off allowed for an unprecedented period of reflection and quiet, and although I remained severely depressed, my anxiety eased, and I felt refreshed upon returning to lab after Thanksgiving. I was optimistic when I returned, and thankful to be back at work, I temporarily held myself to a sustainable pace.

However, in the new year, it became clear that this interlude did not solve some of my underlying weaknesses. My research progress in both labs stalled while I tried to split my time between projects. My programming had improved only marginally, which caused anxiety. I felt my colleagues growing impatient with my lack of progress. Furthermore, Vpr and hHR23A were proving recalcitrant to purification, and a viable path forward for the project was muddled at best. Meanwhile, the lab manager had taken to telling my colleagues that I was too stupid to work at the University of Pittsburgh. Frustrated and bitter, I again considered quitting, but instead doubled down on my efforts, and studied for and passed my comprehensive exams, which is a central milestone in a PhD program. This success provided a brief boost, which was augmented by the news that I had secured an NIH grant for my studies. Still, I saw the writing on the wall- I could not get through this PhD under the co-mentorship. Therefore, after careful consideration, I abandoned my experimental work to  solely focus on computation. I believed that I would be happier in my day-to-day work, away from the lab manager who undermined me, and that I would make faster progress in setting up simulations and writing programming scripts.

Initially, this decision paid off and I made rapid progress. However, by December 2019, I realized that no matter how hard I work, I will never be a PhD-level computationalist. To obtain a PhD, one must deeply understand their research, master the techniques, and develop a unique creative vision for the direction of their project. I was simply not able to make this come together. Therefore, I made the decision to write up the work that I had already completed and drop out of school with a master’s degree. Persisting in my studies would have been nothing more than a vanity project. After two and a half years, my dream of a PhD officially came to an end.

Accepting the unraveling of my degree has been a grieving process. At first, I was in denial, believing that a short break or switching labs would help. After this did not work, I experienced unremitting anger and self-hatred that lasted for months. I was rarely upset with the people around me. Instead, I felt as though I had failed everyone who was foolish enough to believe in my or show me any kindness, patience, or love. My sense of self worth was gutted. In academia, people will pay lip service and say that failure is not a referendum on your worth as a human being. However, when your research is a lifestyle that cannot be compartmentalized or shut away on weekends, when the pursuit of knowledge is your raison d’être, you realize that only those who are successful can make this claim without doubting the veracity of the sentiment.

As I worked through my anger and raw emotion, I have finally begun a journey of acceptance. I am at peace knowing that I will never become a professor or run my own lab. With failure came the opportunity to reinvent my career in a way that I had never thought possible. In retrospect, the worst part of this experience is how my PhD took my love for research and slowly turned me coldly indifferent so that I could survive day-to-day. I lost an important part of myself that I need to get back.

Shortly after leaving school, I was lucky enough to land a position as a research technician. I have never been more thankful that someone was willing to take a chance on me and that I now have the privilege of researching Alzheimer’s Disease. My hope is that by returning to the intellectually stimulating environment of a lab and contributing to a project, I will slowly regain my passion for research that I knew at Clark and UMass.  I can only achieve this revitalization in a job that can be compartmentalized, unlike a PhD. The rhythm of clocking in and out while still being in a research environment has given me a new toehold in science that I plan on sticking to for dear life.

While I am tempted to place all of the blame on myself for failing in my PhD, I realize that the reality is more nuanced. On one hand, my shortcomings- pure ambition, overzealousness, a tendency towards depression and anxiety, and a series of poor tactical choices that landed me in the wrong line of research- led to this outcome. However, there was also a wider programmatic failure that is present in many corners of academia. For instance, Pitt is flexible with students who experience a nervous breakdown because it knows this happens regularly. However, the administration is actively antagonistic to changing the work culture, improving employee benefits, and has litigated for years against graduate student unionizing efforts. A “survival of the fittest” attitude exists among many faculty and training staff. A very large power differential exists between students, professors, and the university. Poor behavior is very rarely reigned in and toxic work environments are frequently tolerated. Pay does not keep up with cost of living increases. There is no guaranteed vacation time or sick leave; this is up to the sole discretion of one’s adviser. I was lucky in that I had supportive advisers who encouraged me to make the right choices for my career rather than simply use my labor to win grant renewals. Their dedication is the reason why I do not want to leave science completely. Nonetheless, if the overall dynamics of higher education do not change, graduate students will continue to burn out at ever increasing rates. My story is by no means unique.