Read our COVID-19 research and news.

Choosing a Thesis Lab


Choosing a thesis lab and a mentor is probably not the toughest problem that you will ever face, but it is an important one, and you will want to get it right the first time. For many of you in combined-degree programs, this will be your first chance to develop a long-term relationship with a faculty member of your choice. The ideal thesis mentor should be successful as a scientist, experienced as a mentor, and willing to commit the time and resources needed for you to become successful yourself. Your relationship with your thesis adviser will be most intense from the time you enter the lab to the time you depart with your thesis completed, but it probably will last even longer, as you call upon them later for career advice and letters of recommendation for future positions.

So Many Labs, So Little Time

OK, it is an important relationship and you need to choose wisely. How do you do that? Depending upon how they are organized, graduate training programs may present students with anything from a dozen to a few hundred faculty members who are approved to be thesis mentors. You may have encountered some of them in class, but most will appear merely as names on a Web site.

This essay is intended to provide some generic guidelines based on my experience with combined-degree students at the University of Pennsylvania and on my own experience as an M.D./Ph.D. student. The advice should apply fairly well to trainees in most M.D./Ph.D. programs offering graduate work in the biomedical sciences. With some exceptions, it should apply to students doing only a Ph.D., but it is written with future physician-scientists in mind.

Students in M.D./Ph.D. programs are faced with a different set of pressures than most Ph.D. students. Chances are, you will do several years of postgraduate clinical training as well, and you are more likely to end up in a clinical department than in a basic science department. Chances are, you will achieve your first job as an assistant professor (if that is your goal) with fewer years of research training than your "Ph.D.-only" compatriots, who typically do a couple of extended postdocs of three or more years before reaching the point where they can compete for a faculty appointment. Their research training will be continuous; yours--as an M.D./Ph.D. trainee--will be interrupted by the return to clinical training after you complete your Ph.D. All these factors make the choice of a thesis lab more critical.

First Principles

A good place to start your search for the ideal mentor and the perfect lab is to consider what the goals of doing a thesis are, or should be, beyond the obvious fact that you have to do one in order to earn the Ph.D. After all, why not just get an M.D. and hope for the best? Reduced to essentials, your goals in doing a thesis should be:

  • To learn how to ask interesting and important questions,

  • To learn how to translate these questions into experiments that will produce verifiable results,

  • To learn how to critically evaluate the results of these experiments, and

  • To learn how to present your ideas and data to the scientific world in written and oral form.

  • Much of the rest follows from these simple goals. You want to be in a successful lab where you can get good advice and work with other scientists and trainees who will help you learn how to do good work. The project you work on should be one you care about, but its value as a training vehicle is at least as important at this point in your training.

    Frequently Asked Questions

    When I meet with first- and second-year M.D./Ph.D. students to discuss the choice of a thesis mentor and topic, certain questions always seem to come up. Is it better to be in a big lab or a small lab? Should I pick a lab where other combined-degree students are already working? Should I try to work for a physician-scientist, or is a Ph.D.-only basic scientist just as good? Should I pick a lab or a topic (or both) based on what I think I want to work on later? Should I avoid working with a junior faculty member? Should I pick a project that I am sure will work (if such a thing exists), or should I pick one that is riskier? How long should it take to complete my thesis? Who decides when I am done? Should I seek, or avoid, labs that do "translational" research? Is it better to work with someone in a basic science department or a clinical department?

    Frequently Offered Answers

    Choose a successful lab. If the lab has been successful before, it probably will be again. If it hasn't been successful in the past, it may not be in the future.

    How do you measure success? Successful labs publish frequently in good journals. Successful labs are frequently successful in winning research grants. In successful labs, previous students--if there have been any--have successfully completed their thesis projects in a timely manner.

    Choose a lab that is committed to you. Pick a mentor who will be committed to your success as a trainee and spend the time that's needed for you to achieve your goals. Although you will be working independently most of the time, it is reasonable to expect your adviser to set aside time for you on a regular basis. If you pick a lab with a successful PI who travels a lot, make sure the lab is well staffed with other experienced researchers who will help you when your adviser's away. Make sure that you have a pipeline to your mentor--and a support structure--for times when you need it. Big labs can be great places to work, but it is possible to get lost in them; will they know you are there and provide an adequate safety net?

    Understand your learning style. Some people work best in crowded, noisy environments; others work best with fewer people around. Some people like to have their supervisor around most of the time and talk to him or her several times per day; others prefer infrequent encounters. Some people like to have immediate access to the lab director; others prefer to wrestle with problems on their own before finding help. Any of these approaches can work, but make sure that what works best for you is in sync with the way the lab you are considering operates.

    Someone else's best choice may not be your best choice. With so many choices available in most programs, it is not surprising that combined-degree students decline to "boldly go where no one has gone before," choosing instead to go to a lab where there are other combined-degree students. Be careful about doing that. The best choice for someone else may not be the best choice for you. Pick wisely, and not just because others have made that same choice. Do a bit of self-analysis, applying the principles described elsewhere in this article, and choose a lab that's right for you.

    Basic versus translational research. This is an issue on which you may get different opinions. Some feel that the best training in research comes from a "basic science" project. Others think that a natural niche for a physician-scientist is to do translational research, so why not start that while doing the thesis project? The critical issue is whether the lab and project will allow you to meet the goals of doing a Ph.D. This can happen in both basic and translational research settings. So ...

    Pick a problem that interests you. You will be living with it for a long time. Make sure it is something you will want to wrestle with even when the going gets rough. It has to make you want to get up early, work late, come in on the weekend, and think about it in the shower.

    Junior faculty members can be great, but make sure they are up to the task. Working with someone who has recently completed training can be exciting. The lab will be small, and he or she will have more time for you. You will represent a larger fraction of their workforce and will, consequently, be valued more. On the other hand, a recent faculty hire will not have had a chance to accumulate a track record as a trainer of graduate students. Your higher value to the lab can also be a disadvantage; you may be asked to carry too much of the responsibility for your adviser's future success or failure. He or she will not have had a chance to prove that they can match their success as a postdoc when they are doing it completely on their own; not all will succeed as independent investigators.

    If you are going to commit three (or more) years to working with him or her, you want to be sure that they will be there with you until you finish. Few things are as difficult as having your thesis adviser not receive tenure and be obliged to look elsewhere for an adviser. So if you are considering working for a new faculty member, look hard not only at what they did while a postdoc and graduate student but also at what they have done (or plan to do next). It is OK to ask about funding, if only because they are going to ask to help support you. It is OK to ask when they are coming up for tenure. You have a right to know these things.

    Senior faculty members can be great, but make sure they are up to the task. Senior faculty members are typically associate professors or professors. In schools where tenure can be granted, they will probably have tenure. A decision to work with a more established investigator avoids some potential problems that can arise with junior faculty members but can raise others. At the very least, they should have a track record by which to be judged. But make sure that the success is ongoing. Look them up in PubMed. Are they still active or just coasting?

    Your thesis topic need not be in an area that you are planning to embrace forever. Most people don't know what they will be doing in the future; chances are that it will change over time. However, make sure it provides you with opportunities to gain the skills you need to be successful as a scientist.

    Avoid "Ph.D. lite" (a.k.a. "combined-degree Ph.D."). M.D./Ph.D. students understandably feel pressure to keep moving through the various phases of their training. Don't succumb to the temptation to do less than your best. Give yourself the time you need. Set the tone for your career as an investigator by choosing challenging questions that might take a while to answer rather than just doing incremental studies that are designed to be finished quickly. In the long run, you will be far better off.

    Be thoughtful in choosing a thesis committee. Pick people who are experienced in training graduate students as well as expert in the area of your research. The chair should be someone with sufficient weight to act as a counterbalance to your thesis adviser should any difference of opinion arise. Meet with the committee regularly to avoid getting bogged down when things aren't working. ("This experiment didn't quite work the way I wanted it to the first 87 times, but I am sure that if I do it once more it will.") Proper use of a well-chosen thesis committee not only gives you a chance to practice your skills at presenting your work, it also helps you use your time as efficiently as possible.

    How long should it take? As long as it takes to accomplish the goals you set out, and to accumulate an identifiable body of creative work. Most combined-degree students take 7 to 8 years for the entire program. Taking longer can be fine. The key is to be focused and efficient and to work hard. Your thesis committee in consultation with your thesis adviser will decide when you are done. The decision is usually based on evidence that you've mastered the field, achieved the goals of doing a thesis, and made a scholarly contribution to the field. Most programs strongly encourage M.D./Ph.D. students to complete and defend their thesis before they return to medical school to complete their clinical rotations.

    Learn how to recover from failure. Success feels great, but you must also learn how to recover when things don't work. I've known combined-degree program graduates who sailed through their thesis research only to crumble when faced with bigger challenges when they returned to the lab after completing their clinical training. Suddenly, being a full-time clinician seems much more attractive than wrestling with experiments that typically fail the first time around.

    How Do I Start?

    When you arrive on campus, establish a habit of attending events in your department or graduate group. Chances are the training program will have an event in which faculty members describe their interests; go. Talk to the faculty members who you are considering. Don't be shy; this is an important relationship for you and for them. Ask critical questions about their commitment to training you, their success at training others, and (in general terms at the least) the status of their funding for the period you will be working with him or her. Ask junior faculty members how soon they will be up for tenure and whether they think they will get it. Ask if there is a chance that they might leave the program while you are still doing your thesis research. Look at PubMed and make sure that they are publishing regularly in quality journals, including Cell, Science, and Nature. Read their publications. Talk to the people who are working in their lab. Find out what really goes on. Tap into the collective experience of other M.D./Ph.D. and Ph.D. students who worked in the lab. If possible, "try before you buy" by doing a lab rotation. At the very least, go to the lab's weekly meeting and listen to the discussions. How much give-and-take is there? Do students seem engaged?

    Final Thoughts

    Many of us who have completed training look back on the years when we were thesis students with fondness. No doubt, this is partly a result of imperfect memories; anxieties and stresses fade over time. Nevertheless, this is a special time. This will be one of the rare periods in your life when you focus all or nearly all of your energy on research. Take advantage of that. Choose your thesis lab and thesis mentor carefully, and enjoy the process as well as you enjoy your accomplishments.

    Follow Science Careers

    Search Jobs

    Enter keywords, locations or job types to start searching for your new science career.

    Top articles in Careers