All Articles

Doing a PhD in 2022

So I got into grad school. One of the things I haven’t really put a lot of thought into is the mechanics of being a PhD student. What do people do on the day to day? What activities are people involved in, which skills should I try to learn, how should I interact with other students and faculty, etc…

There aren’t a lot of guides out there. MIT has a cool blog series written by current students. Specific to the day to day of being a scientist however, I couldn’t find anything. At the end of the day, I want to be a world class scientist. So, this post is summarizing all of the advice I’ve received about doing a PhD and a review of major technologies or activities that I thought might be necessary as part of graduate training. This also serves as a time stamped document of my naïvity so that I can look back in 5 years at how stupid I was.

General Topics

Graduate school is technically training, but to some it is simply work. We input time, energy, and a massive pay cut to get prestige, publications, inventorship, classes, and a suite of soft skills. I think too many people poo-poo on graduate school, and for some rightful reasons, but I’m hoping that just like college or any other job, it is what you make of it. For me, I want grad school to do the following:

  1. Allow me to publish at least 1 Cell/Nature/Science level paper as first author
  2. Invent at least 3 patentable technologies
  3. Be able to collaborate with anyone and manage relationships with PIs/advisors, peers, undergrads, collaborators, people in industry etc…
  4. Give me the opportunity to consult for at least 3 startups and intern at a venture creation studio
  5. Give me the skills to properly design studies, plan projects, and manage a team
  6. Build a knowledge base for me to be the smartest person in the room 99% of the time
  7. Be able to design beautiful slides and presentations and write clearly and concisely
  8. Gain technical skills to operate equipment and carryout experiments which I will describe below

I think its important to ask at the end of your graduate training, what will be your unfair advantage? I don’t think you need an unfair advantage for all post-PhD careers (for some careers the presence of a degree is already an unfair advantage), but for some you do. Competitive career paths like medicine, venture capital, and entrepreneurship require an edge and a reason to build conviction in you as a person over anyone else in the world for the job.

I think there is a lot of learned helplessness that happens during grad school. Much I guess is deserved since straight out of undergrad if you go to a program that isn’t supportive of graduate students it can be overwhelming. However, for top students at top programs, I think you should be able to be in the driver’s seat for the majority of your time there. Having the ability to be resourseful, ask for help, and make progress in the face of adversity is something that you just have to be able to do no matter what in whatever you do. I’m a believer in the value of graduate training, especially at elite institutions very much for this reason.

Why do a PhD?

The PhD at its core teaches you how to invent, to go from 0-1. Its a professional degree that gives you deep domain expertise, the ability to collaborate and work in a team, and connections to top scientists.

A PhD certainly opens up doors in scientific organizations like biotech/biopharma companies. If you ever want to lead and recruit a team of scientists, it makes sense for you to have gone through the same path they did. A PhD is a badge of honor, demonstrating your capabilities as an independent thinking and innovator. I’m sure you can progress and earn this badge without doing a PhD, but its much more difficult.

A PhD is flexible. You can work on whatever you want and choose an advisor and peers that suit you. You become a deep domain expert, making you valuable as an advisor and thinker. During your 5-6 years in training, you get to do internships and manage your time on your own schedule, unlike any other job.

A PhD gives you quite simply an excuse to tinker and play with science without major implications if you fail. I’d like to set up a home lab one day and conduct my own experiments at home (make my own GMOs, biomaterials, etc). You can’t learn how to do this from the internet.

I don’t think I’m ever going to do a post-doc, so I’m going into my PhD thinking that I’m going to cram it all into the 5-6 years that I have. A big bet on myself. I think its always important to ask “What do you want out of a PhD?"" Is it papers, citations, and awards? Is it patents, internships, and a job? For me, I’m planning on going to industry or medicine after my PhD and definitely do not want to become an academic researcher post graduation. As a result, I don’t care about citations or paper count, but rather how impactful my work is translationally. These can potentially be measured with patents, experiential learning experiences like internships/shadowing, and maybe awards.

Time Management

In terms of hours, can you really treat it like a 9-5? I’m sure we’d all like to try but timepoints will require weekends and nights. Ideally, I’d work around 60 hours a week, spending the rest taking care of myself and making sure my other needs in life were fulfilled. But what I love about science is that you can essentially set your own hours. You have a PI who in some sense is your boss, but you can decide how often you want to meet with them and ultimately some of the stuff you are doing is impossible so hate to say it but there are always excuses. In this sense it is quite relaxing and you can take vacation whenever you want. The issue is everyone wants to graduate and most people in these programs are self-motivated enough to feel bad about not working enough hours. But this flexibility is a major plus especially in today’s Zoom centered world where you can fit so much more into a day. It makes it so much easier to do consulting projects, workout, eat healthy, and find space for other activities and errands.

The downside is that there is little separation between your work and life. You always can be reading papers or working on your research, especially if there are computational components of your project. Having planned trips or hangouts might be a good way to mitigate this, but also being good at not letting work interfere with your personal alone time.


Along with raw counts of hours working in the day, the best PhD students work efficiently to not waste time. A couple tactics I thought were cool:

  1. Telling people exactly what you want from them. One grad student I know gave their undergrads typed out instructions for each day that they came in. In emails, formatting them clearly and concisely and not worrying too much about politeness.
  2. Proactively make slides to describe motivations and come up with an explicit mathematical hypothesis to frame results.
  3. Reading, talking to people, and other background research will save you time, money, and face
  4. Basic operations like figuring out rate limiting steps and doing those first

Pretty simple stuff.

Has the PhD changed in the past decade?

I’m a believer that the PhD has changed significantly in the past decade and there has never been a better time to be a graduate student than right now. Sure, in a lot of areas, some lower hanging fruit has been picked. However, we have so many tools now to discover new things and arguably if you have the money and time to investigate something, you can almost guarantee a high impact publication. For example, in the past these might have been solving crystal structures of certain proteins. More recently, it might be using Cryo-EM to do the same thing for hard to crystalize proteins, directed evolution of a really useful protein, single cell sequencing and analysis of a bunch of clinical samples, a large microbiota study, “Atlas” studies, or any of the CRISPR screen identifies “X” studies.

In other words, there are plenty of 1->100 papers to be published. These can sort of become a hedge in case your 0->1 projects don’t work out. And when the 0->1 projects do work out, there are almost always high impact.

There are also so many more tools to collect data with. The cost of sequencing has dropped, the cost to synthesize probes and oligos has dropped, the number of startups each with their own special technology has skyrocketed. We now have spatial -omics, nanopore sequencing, and an impossible number of computational tools. CRISPR and AddGene have made molecular biology simple and cheap. So many more questions are answerable by the humble graduate student which before could only be done by large teams.

Take the number of diseases, biochemical pathways, and phenomena and then multiply by the number of new technologies and study them, and then multiply again by the number of ways to perturb the underlying biological system and we have millions of potential high impact papers. We have so much to explore. The rate limiting step is our ability to ask intelligent questions, communicate results, and collaborate to repeat the sceintific process over and over again. It is an incredibly exciting time to be a trainee.

The downside is, you have to seek out mentors and collaborators to teach you all these new skills. This can take a long time and people can be slow or unable to help. This is why I think its important to join a ‘hub’ where many scientists co-localize. It reduces the friction of asking and getting help many fold.

Another thing that has changed for the better is the wealth of resources for graduate students enabled by the internet and accelerated by Covid-19. AddGene, bioRxiv, YouTube, Twitter, community Slack channels, and a wealth of free online seminars have made learning much easier. Not only does it make science easier, extracurricular activities and cross institution collaborations make student organizations that much more powerful. The success of the biopharma/biotech industry has also catalyzed student entrepreneurship and participation in the venture ecosystem that I think is quite rare.

What new tools should all new bioengineering PhD students understand and know how to use?

This might be somewhat controversial, but I think ideally as a super scientist you should be fluent or at the very least knowledgeable about all related diciplines and techniques. This is one area where I think VCs are pretty good at. They get a nice birds eye view of many separate fields and are able to develop perspectives on how technologies may interact and show synergy, and also where there are areas of need. I’m someone who wants to lead teams of scientists and develop new drugs and I think this is exactly the right type of mindset to have for this role in the ecosystem. Developing an understanding of all the different types of therapeutic modalities, how they are studied and developed I think will be quite valuable.

For people that say this is impossible or unnecessary, I disagree on part 1 but partially agree on part 2. It isn’t impossible to learn quickly and it never was. There are network effects in learning as well, and once you have an internal map of how different fields in science and concepts interact, it isn’t that difficult to fit things in. To make things even more efficient, for most papers, you can skim the most important sections, you can also set up calls with authors to learn super quickly what might take you several hours to digest from a paper. Finally, you should constantly trade notes with classmates about what they are learning about. Its super important to have peers with similar goals to you, or at the very least the ability to talk and share perspectives and ideas. I agree however, that this approach isn’t for everyone. There is absolutely value in learning different fields and how concepts in one can be applied towards the other — there is no doubt that this accelerates research. It is a tradeoff however, and some people can get overwhelmed with the prospect of spending significant time outside their home field. There is something to be said about intense focus as well.

Below is a list of technologies I think will potentially be really useful one day (and ones I hope to learn during my PhD):

  • Next Generation Sequencing: including Illumina and Nanopore sequencing, library prep, data analysis, probe design, sample prep
  • Lab Automation Software: both proprietary and self assembled. I want the ability to program any piece of equipment whether software exists for it or not yet. If rigs need to be built to automate things, I want to learn how to do it
  • Flow Cytometry/FACS: an essential tool for immunology, you can’t not know this. Included here is FlowJo
  • Microscopy: I want to be able to take beautiful pictures and visualize structures at the nano and micro scale. This includes TEM, SEM, confocal, fluorescence, and all the prep, reagent considerations, and technical details
  • Cell Culture: being able to play with cells and keep them healthy for experiments. Primary, co-culture, organoids, etc.
  • Animal Studies: Conducting experiments with mice at the very least and modeling PK/PD, ADMET, and efficacy in animals is necessary for developing drugs
  • Advanced Histology: CODEX, cryosectioning, handling of clinical and animal samples.
  • Software Development: R and Python for scripting, RShiny for dashboards, basic machine learning, plot generation, basic statistics, git, sharing reproducible code
  • Bioinformatics: the ability to use NCBI tools (Cobalt, BLAST, accessing sequence databases), accessing and using computing clusters, and ability to use common software like SnapGene, Geneius, Benchling, and the ability to employ published packages, models, and software for my own applications
  • Structural Biology: NMR, Cryo-EM, PyMol, Mass Spectrometry
  • Chemistry: How to do syntheses of small molecules, peptides, etc. How to purify compounds with chromatography
  • Biomaterials: transfection reagents, cell seeding matrices, building homes and other environments for cells to interact with
  • Molecular Biology Activities: Westerns, PCRs, CRISPR, cloning, transfection, screening, primer design, directed evolution, basically how to play with DNA, RNA, proteins, and cells safely and effectively.
  • Microfludics: Design of customized devices with CAD

Clearly I don’t know what I’m talking about for some of these but alas. I think its important to be extremely tech savvy during your PhD to the extent where you know how all these machine work, what the limitations are, and what new machines in development might help your research.

Futher Reading

Nature Communications - Understanding the onset of hot streaks across artistic, cultural, and scientific careers

Rishi Kulkarni - Asking the Right Questions During Your PhD

Published Feb 20, 2022

Harvard-MIT PhD Student | Biotech Consultant