Unto the breach: Navigating a research pivot during your PhD

Erin Berlew

So it’s come to this. Maybe you’re staring down some experimental results that sink the hypothesis that’s the premise of your work. Or maybe a conversation with a thesis committee member made you realize that there’s a major, no-coming-back-from-it flaw in the logical foundation of your project. Or maybe sequencing results show that what you thought was a cool new protein is actually just contamination in your sample. However the need to pivot presents itself, I want you to know three things: (1) This happens to everyone at some point and usually more than once, (2) it’s okay to have feelings about it, and (3) you will come out a better scientist on the other side.

​I’ve experienced the need to pivot scientifically a few times. I’m a fourth-year PhD student and my project focuses on using natural photoreceptors to engineer new optogenetic tools. In the first year of my program, I inherited a project that had been on the lab’s backburner for a few years. In theory it should have been really cool—characterizing a new photosensory protein and using it to catalyze intracellular reactions. A few students before me had worked on it and couldn’t quite get it to work, and after months of trying (spoiler alert!) neither could I. The second pivot point came when I thought I had found an exciting new fluorescent photoreceptor which turned out to be the result of contamination in a communal media stock. Both experiences were bummers in different ways, but both taught me how to change direction, deal with the feeling of failure, and grow as a researcher in the process. In this post I’ll take you through the lessons I’ve learned when projects hit a dead end and provide a roadmap to help you navigate the pivot process.

During a research pivot I visited my friend Lindsay in Pittsburgh and did some brainstorming in her coworking space. A change of scenery + quality time with a college friend = a productive break from the lab.

​ Identifying a dead end

How do you know a project has failed, or at least needs serious rethinking before more progress can be made? Sometimes the decision to pivot is straightforward: the project has crashed and burned. You find a paper that’s accomplished exactly what you were trying to show, or a well-designed experiment pokes a hole in your foundational hypothesis. For me, it was sequencing results. A persistent plasmid contaminant had found its way into my glycerol stock, using the same promoter and terminator as my gene of interest, making it difficult to spot with sequencing alone. Eventually, after many sample preps and bacterial colony isolations, I figured it out, or the sequencing company did: the impressive phenotype I thought was coming from a previously-uncharacterized photoreceptor was actually contamination by a well-known fluorescent protein. It was an unpleasant result to receive by email, but it was clear that the path forward would have to change.

Other times, projects fizzle out over time. While less dramatic and acutely stressful, slow-burn project collapse can feel worse: it’s accompanied by longer-term, low-grade stress that something isn’t working in the lab, and it forces you to make the decision to pull the plug on an idea you like that you’ve invested a great deal of time and brainpower into. When you know you’re close to project breakup but aren’t sure if it’s worth two more weeks of work, here are some questions to consider:

1.     What would need to be done to get this project to its goal?

Write down the experiments or findings that would push the project past whatever its next finish line is (a publication, conference presentation, thesis chapter…). Then assign values to these steps. How long would it take you to complete six more rounds of directed evolution on your catalyst? How much of your training grant would be eaten up by the cost of sending out the samples for mass spec? Importantly, what is continuing to work on this project preventing you from doing at the same time? It’s possible that your project is closer to the finish line than you thought, or that there’s no clear way to get it over the hump. Either way, concretely laying out the project’s direction will give you clarity as you decide whether and when to move on.
​
2.     Am I equipped to carry this project to its next endpoint?

Once you’ve identified the project’s next steps, think about your experimental and analytical skills as they relate to these steps. Does your research right now require extensive statistical analysis when you’ve only taken an undergrad stats class? Are you ready to compare thousands of fluorescence spectra but your lab only has a single-cell spectrophotometer? Naming the disconnect between what your experimental plan calls for and the resources you have at your disposal will help you decide on a path forward. Could you get a postdoc to help you fill in the gaps in your knowledge, or pause on the project until after you’ve taken next semester’s course load? You also may decide to punt on the project until your lab has higher bandwidth. Looking for gaps between your capacity and your project’s needs can keep you from running into an experimental wall.

3.     Does this project as it exists align with my personal and professional goals right now?

Ask not only what your project needs from you, but what you need from your project . It may be that you’d be ready to publish if you devoted six months only to mass spec. That might work for your trajectory as a student, or you may decide that such a narrow focus will prevent you from learning other skills you want to get out of your PhD. Think about your graduation timeline, too: will finishing the project as it’s designed allow you to defend on a schedule that works for you? Research projects are a two-way street; you should be getting value out of what you put into them.
 

Communicating the need for change

Once you’ve identified that you need to pivot and change what you’re working on, it’s time to get your adviser on board. I’d recommend talking to your adviser as soon as you can, both to reduce your personal stress and to keep everyone on the same page for conference talks and grant writing. If you have a quality research mentor, this conversation can be extremely helpful in deciding where to go next and reflecting on the project you’re leaving behind for lessons the lab can learn. If you have a Tormentor (the Double Shelix podcast’s brilliant term for substandard PIs), this conversation can be intimidating.

I want to remind you that research failure is not personal moral failure: unless you intentionally set your lab bench on fire or falsified your data, you haven’t done anything wrong; you’re just experiencing a scientific rite of passage that your adviser has also gone through. In many cases, especially early in your PhD, you might not have even chosen the project you’re working on in the first place! You’ve taken the project from the “interesting idea” phase to the “we learned something” phase, even if what you learned is disappointing, confusing, or not immediately publishable. Much as I, a cat owner, dislike the Schrodinger’s Cat paradox, you’ve opened the box and learned the truth, which is an essential and non-glamorous part of the scientific process.

While I’ve been lucky in the mentorship department, I know that the step of talking to your adviser about a failed project is probably the most difficult and stressful of the moving-on process. If you’re struggling to explain to your adviser why your project isn’t working anymore, here are some ways to frame it:

• The project needs X technical/mathematical expertise that we don’t have; outsourcing it would be expensive.
• Continuing as planned would put us Y months away from being able to publish what we have, which is much longer than we initially planned.
• By not working on this project, it would free me up to start on project Z, which we agree is more impactful.

If none of these rationales appeal to your adviser, it’s time to get advice from others who understand your situation, namely senior lab members or other faculty. Other grad students or postdocs in your lab have likely had to break bad news to the boss before—what suggestions do they have for framing the conversation? Depending on lab dynamics, maybe one of them can also talk to your adviser to give their perspective on your project and how it’s no longer the best use of collective time and resources. If needed, get your committee involved —sometimes faculty will take the tough news more seriously when it comes from other faculty. This can be especially useful when thinking about how a dead-end project could negatively affect your graduation timeline. If you don’t yet have a committee, consider talking with your grad group chair or any professor you have a positive relationship with. They can help communicate your needs as a student to your adviser, which are just as important as the direction of the lab. Whoever you consult, having someone on your side as you navigate this change in research direction can help alleviate anxiety and make a plan of action.
 

Wrapping up a failed project

Perhaps the most frustrating part of stopping work on a project is tying up the loose ends. You’ve made the difficult decision to pivot, you’ve gotten your adviser on board, and maybe you’ve even gotten excited about a new idea, but you still have to spend more time thinking about a project that isn’t worth the time?! When I got my contamination sequencing results back, the last thing I wanted to do was spend more time and energy looking at my notes and going through old samples. In the long run, though, I’m really glad I did: I ended up being able to use some of my DNA primers to prep other genetic constructs down the road, I’d written protocols over the course of characterizing “my” protein that were really helpful for other experiments, and it gave me some closure as I looked back at all the work I did, even though it didn’t come to the end I’d hoped for. Even if you’re sure that none of your work on the old project will benefit you in the future, it’s possible that another student will come back to it years from now when the lab has a new piece of equipment.

Keeping good records is part of good science, and spending a few hours closing out a project is a public service for future lab members. In fact, the Dark Reactions Project developed a machine learning algorithm to predict more efficient chemical syntheses from the unsuccessful chemical reactions detailed in student lab notebooks! Who knows where your work will pay off?!

As you wrap up your old project, here are some steps to consider:

• Write a brief project narrative. Summarize the goal you were trying to accomplish, any key results from experiments you did, and why the project ultimately didn’t work. This document could prevent a future grad student from being disappointed by the same idea down the road!
• Combine your data presentations into one PowerPoint file for easy review later on.
• Collect any biological samples (DNA, proteins, cell lines) and make sure they are appropriately logged and easy to find. Note their storage location in your project narrative document.
• Back up your data to an external hard drive and/or the cloud, then feel free to ritualistically delete duplicate files in a moment of catharsis. 

 

Assessing your skills

While you’re going through files and materials from your old project, you will probably come across evidence that even though it may not feel like it, you almost certainly learned something over the course of your work. Challenge yourself to make a list of good things you’re taking away—to get started, you definitely improved your ability as a scientific decision maker in figuring out you needed to step away. What “hard skills” do you have now that you didn’t have before? Can you do a pulldown assay, work with primary cells, or code in MATLAB? How about types of analysis or computer programs you can now work with? If you’re mad at your old project, you can even list things you wish you’d known before, like how to detect contamination for samples in similar plasmid backbones.

While it was really difficult to walk away from projects, there’s no question that they changed me as a scientist. Trying to characterize a “novel” protein taught me to interpret macromolecular mass spec; even though there was nothing novel about that protein, I still know more about how proteins fragment in response to degrading agents. Give yourself a gold star for every new thing you can do now, and take them with you as you move onto greener pastures.
 

Dealing with feelings

 I know that so far I’ve made the process of admitting you’re at a dead end sound pretty clinical, but the experience itself can be messy. Allowing myself to feel sad that a project I’d previously enjoyed wasn’t going well, frustrated that the academic trajectory I’d planned out based on preliminary results was going to change, and angry with myself for not catching contamination sooner were important steps in my process of moving on as a scientist. Give yourself permission to grieve the loss of something you’ve put a lot of yourself into. Know that while you may feel like an impostor, you still belong in science. Your feelings, even the negative ones, are meant to be felt, and pretending everything’s fine will only make your life more difficult in the long run.

Some practical tips for coping emotionally with a project ending:

• Consider taking a few days off from lab and spend time doing things you enjoy. Going to a park, watching trashy TV, or simply spending time out of your routine can help you gain perspective and remember that you are still a person in addition to being a scientist.
• Make sure you’re taking care of your body and mind, whatever that means for you.
• Know that everyone experiences scientific disappointment at some point. Talk to other grad students in your lab, your cohort, or your mentors to hear about their experiences of projects gone awry and take comfort in the fact that no one’s scientific trajectory is linear.

 

Starting something new

Deciding to leave a project and starting over is rarely a fun experience, but it does give you something of a clean slate. You can jump into a new research question, start a new lab notebook, and enjoy the possibilities of an early-stage project. But you don’t have to decide how to spend the next few months or years of your life right away—take time to make an informed decision that’s right for your work style, research interests, and professional goals. Here are some tips to set your new project up for success:

• Make a list of skills you hope to build over the rest of your time in the lab. Look at prospective projects with an eye to how they can help you develop these skills.
• Talk to your thesis or advisory committee. Their job is to help you graduate, so getting their input on your research path will be helpful as you plan out your time.
• Shadow a lab-mate or someone in another lab for a few days whose project is different from yours. What does their day-to-day work look like? What do they like and dislike about the work they do?
• Take a week and catch up on the literature in your subfield, like you did for your qualifying exams except not as emotionally draining! What are your favorite labs working on? What open questions interest you? Where are the gaps in knowledge or technology that you could try to fill?
• Schedule a lab meeting where you can brainstorm new project ideas with people who are intimately acquainted with your lab’s work and wheelhouse.
• Think about when you knew you wanted to be a scientist. What excited you about science then? What parts of being a scientist inspired you? What kinds of research questions or projects give you that same excitement now?

 
In closing, changing course in research is a difficult adventure. Every scientist experiences the need to pivot at some point, and the failure of a project to produce results is not an indicator of your worth as a person or a scientist. If you are going through the need to leave a project and find another, take a deep breath, go easy on yourself, and welcome to the club.

Erin is a PhD candidate in Bioengineering at the University of Pennsylvania studying optogenetics and cell signaling.