My Thesis Defense

T-2 months

I started writing my dissertation after I returned from holiday break. In my program, the dissertation usually consists of three or so papers. They do not necessarily have to be published, but they should be plausibly publishable. Since I had some paper drafts already, when it came time to write my dissertation, it mostly consisted of writing an introduction and a discussion to unify the three papers, plus some additional details like formatting according to the university guidelines and adding references.¹ Then a month before my defense date, I sent a draft to my thesis committee for them to read.²

Around one month before my date, I also started preparing the slides for my presentation. Similar to the dissertation, I realized that because I’d presented a good portion of my work previously (informally in lab meetings or at conferences), I could re-use the slides I had already. I did add more introductory slides to give more context and background to my research, keeping in mind that the audience for the thesis defense is more general than usual; for instance, your thesis committee has to include members outside your department (based on rules set by the university)³ and ideally, you would like to have your friends and family who are attending to be able to understand what you are saying too.

T-1 week

Factually, what I described above was what I was working on in the last month: preparing the slides and practicing my talk. However, emotionally and mentally, the last week or so leading up to the defense were particularly challenging in my experience. (If you are a current PhD student thinking about their thesis defense, feel free to skip to the next section to avoid being infected by my anxiety.)

Perhaps being so anxious was not surprising. After all, the thesis defense is a unique concoction of a stressful event. There’s the fact that with a public seminar, you’ll have friends and family attending, who are probably learning about your research formally for the first time and so you don’t want to disappoint them. Nor do you want to disappoint your advisor or department, since the quality of your defense is not just a reflection on you, but also them. Then there’s the stress associated with the unknowns of the closed door session, the last “exam” of your PhD.

So for me personally, the stress brought out a lot of self-doubt leading up to the thesis defense, maybe more so than at any previous point in my PhD. Just the week before my defense, I learned that I had a complete misunderstanding about a very basic fact related to genomics. That realization heightened my fear that I didn’t know the field well enough to be able to “defend” my research, let alone talk about the broader scope of my work. I would also find myself falling into a spiral of negative thoughts. For example, I might start off thinking about how it’s true that you are expected to pass the thesis defense; in fact, the department treats your defense date as a celebratory event before it even takes place and I have not heard of anyone in my program failing to defend their thesis. However, it’s still within the power and right of your thesis committee to fail you… so the fact that no one has failed their thesis defense just meant that it would be even more ignominious if I were to fail!

The Day of the Defense

Regardless of how nervous or stressed I was, the day of the defense arrived all the same. I logged onto the Zoom session a couple minutes before the scheduled time and waited for the committee members and audience to show up.⁴ Then my committee chair asked my advisor to introduce me, after which I gave my talk for about 50 minutes.

The talk itself was a blur. I’d worried about not being rehearsed enough the night before my defense, but it must’ve been sufficient because I don’t think I had too many stumbles and the words just flowed out, despite me being nervous for much of the presentation.⁵ After I finished my talk, the chair invited questions from the audience.⁶

Then the chair asked everyone to leave for the closed session, which is when the committee members ask me questions, much like the preliminary oral exam. The difference from the preliminary oral exam, with my committee at least, is that the questions here felt slightly less interrogative, maybe because there’s less of an intent to really test you and more of a desire to discuss the research you’ve done.

In my closed session, the discussion was organized by project and lasted for ~45 minutes in total. We discussed the three projects in the order I’d presented them in my defense. To give future students a sense of what it’s like and demystify the experience a little bit, I’ve tried to recall the questions I was asked below. I will call my committee members A, B, C, and D.⁷

Project #1:

• A - It’s estimated that most radiation exposure comes from radon. How do you think that affects your results?
• A - What are your thoughts on radiation hormesis?
• B - If you are using mutated mice, how does that affect your results, in comparison to normal mice or humans?
• C - How would different remaining life expectancies affect your conclusions?

Project #2:

• A - Could you find mutational signatures at the individual level?
• C - How many mutations would you need to observe in order to conclude that a person has a mutational signature?
• C - What are your thoughts on why signatures differ by tissue?
• D - Could you have mutational signatures for different types of mutations?

Project #3:

• C - What is the statistical distribution of bulk RNA-seq counts?
• D - If you were to design an ideal experiment for assessing statistical distributions, what would that look like?
• D - (Follow-up) How would you design synthetic spike-ins?
• C - Couldn’t you evaluate the fit of different distributions using the performance on a held-out test set?
• B - How do you know if the mean-variance relationship is a good enough fit?
• D - What would you say if someone (a non-statistician) were to ask you why we don’t just use the normal distribution to model the data?

After we finished the discussion, I was asked to leave the Zoom meeting for a few minutes. When I returned, the chair and the rest of the committee congratulated me on my successful defense. And that was it! I was done with my thesis defense!

I stayed on the Zoom meeting for about another hour for a virtual celebration, during which other people from the department joined to congratulate me and we chatted about various things. I had not seen some of them in a long time due to the pandemic, so it was great seeing them again. I also got congratulations the rest of the day from my friends outside the department, many of whom were able to attend my thesis defense because it was virtual.