Choosing a PhD

First, should you do a PhD? In my opinion, there are two reasons to do a PhD: if you know exactly what you want to do and if you have absolutely no idea what you want to do. If you know what you want to do, it’s simple - find the best scientists in that area, email them, figure out the funding and get started. Unlike in industry, email addresses of people in academia are usually listed on university websites. If not, you can find the email address of a scientist by looking at the papers of which they are a corresponding author.

If you have no idea what you want to do, as was in my case, PhD is an incredible opportunity to learn how to learn independently, figure things out and truly develop yourself, all while being paid (if you are lucky to have a scholarship). In the UK there are Centres for Doctoral Training (CDTs) and Doctoral Training Programs (DTPs) which, unlike a traditional PhD, allow you to gain a bit of experience before commiting to a years long project; similar options exist in other countries.

Science is all about asking the right questions. The same is true for choosing your PhD project. People will often ask you what you are interested in; you are likely to meet this question on the application forms too. However, this is not the right question to ask when considering your PhD options. No matter what you find interesting today, at some point you’ll get frustrated with the slow progress of your project or with the lack of support from your supervisor or something else. Ask any PhD student, everyone had some days when they felt not interested in anything at all. That is why you should choose a PhD project that you consider important for the future of humanity. This sense of importance will be the driving factor that will motivate you every day, no matter what.

Explore your PhD options widely in terms of academic areas (and geographical ones too). If you find a particular area of research important (and hopefully interesting too) and you are motivated to learn, you’ll be able to make a valuable scientific contribution even if you have no prior training in that specific area. On a selfish level, it also adds value to your PhD as you’ll essentially get a free Masters degree in a new subject or two.

May 2023 update: A book called Towards a Global Core Value System in Doctoral Education was published in 2022 (with open-access PDF). If I was about to start my PhD I would probably want to read it. And if it’s good, I’d make my supervisor read it too.

Starting a PhD

It will be overwhelming - you’ll face a mountain of papers to read and a ton of jargon to learn, and you’ll be under pressure to actually start doing something as soon as possible. It will be tempting to just get stuck into it, only leaving your desk to refill your coffee mug. But trust me, someday you’ll be longing for somebody to talk to about your project, or worse about a problem in your research team. You’ll realise that people outside academia don’t quite understand what bothers you. At that point you’ll regret not making friends in your building early on, and it will be very awkward to introduce yourself to somebody whom you’ve seen around the coffee machine many times and ignored them. Introducing yourself to everyone and making friends is also a valuable investment from a scientific point of view. If you talk about your project and the challenges you face, you’ll inevitably bump into people who’ve solved similar challenges before and will be able to help you, saving you many days of struggle. If you show interest in other people’s projects, you’ll discover which of your skills or knowledge could be useful to them and be able to get involved in exciting collaborations.

Starting a PhD is a challenge, and even experienced professors remember how hard it was, so don’t be afraid of asking naive questions. Get into the habit of googling your question first and, if that doesn’t answer it, then reaching out to a colleague or indeed anyone who might know the answer. Once you get the answer, also ask them if you could have easily found the answer yourself, that is invaluable information to have.

No matter what discipline you are in, your PhD will probably involve using a computer. Even if you are a very non-technical person, there are several skills that you’ll learn eventually. Investing a bit of time early on will save you a lot of trouble and pay off ten-fold. Here is a list of my top five tips in the order of importance:

1. Backups. Everyone believes that ssd/hard drive failure will not happen to them; but it does happen. Losing a month of work is devastating, but even a day of intense writing would be a shame to lose. An easy way to get yourself into a good setup is to work on your personal device for a day, then work on a university computer for a day, then work off-line for a bit. That ensures that you’ll have a copy of all your files backed up on a cloud and also available on at least one physical device. Some universities offer hourly backups, but they have hardware/software failures too. Having all of your work backed up by both your university and a private provider (like Dropbox that is built around having a local copy of files on your device) is ideal. It might take a while to set it up just right, but it’s totally worth it.

2. Keeping notes. Do you regard yourself as somebody with a good memory? Doing a PhD will show you your limits pretty soon. You’ll be bombarded with a colossal amount of information - from the papers you read and seminars you attend, from experiments you do, and ideas your brain generates. “I don’t need to write this down, I’ll remember it” is the biggest lie you can tell yourself. Some things are just too obvious, you won’t write them down, and you’ll forget and eventually you’ll learn to write everything down.

   The most important feature of your notes is that they are searchable, so keeping typed notes is better than a handwritten notebook or an audio log. Ctrl-F shortcut is the best thing ever, but sometimes you might also need some regex. That is a good reason for keeping your notes in a plain text format (perhaps in markdown syntax) using a text editor (Sublime is my favourite). You might be comfortable with keeping your notes in Word/Google doc, but having a single file thousands of pages long is impractical from a technical point of view, so you’ll definitely want to split it into different files. That creates a problem, as searching across several Word documents or a folder on Google Drive is not easily impossible. Plain text files (all of them at once) can be easily searched with command line tools like ripgrep.

   Another good reason to use plain text formats is that they can be version controlled with tools like git (and tools built on top of them like GitHub). Have you ever had a folder with MyDocument, MyDocument-version2, MyDocumentFinal, MyDocumentFinal-withComments, MyDocumentFinalFinal, MyDocumentFinalFinal-version2? Investing your time in learning how to use a version control tool will definitely pay off in the long run. Once you fall in love with plain text formats, you should try LaTeX. It might seem like a lot of work to learn it, but once you get to the stage of writing your hundreds of pages long thesis you’ll be so grateful to yourself for learning it early on. I’ve seen a face of a PhD student who wrote their thesis in a Google Doc, added a couple of final paragraphs a day before the submission, and then tried to open the document to print it for submission only to see “Cannot open: document too big” warning. Writing your thesis in a what you see is what you type format has all sorts of benefits, it would take a whole book to describe them all.

   You’ll probably end up with a lot of lists - a to-do list, a list of ideas to try, a list of papers you found on a particular topic, a list of other labs working in your field, etc. Putting all of these lists into WorkFlowy made my life much easier. You can keep both PhD and personal notes there, share parts of your notes with other people (in view-only mode), collaborate with other people, access your notes from any device by simply logging in through a browser and also work offline (using a desktop app).

   Three types of notes you should probably keep even though it might not seem like a good use of your time at first:

   • Notes about every paper you’ve read, including an easy way to find that paper again. At first it seems that you’ll definitely remember what the paper was about, and even if you do forget you can always simply read it again. In the long run, it will save you a lot of time to have short summaries of relevant papers in a searchable format.
   • Notes about everything you did that didn’t work - an experiment you did, a piece of code you wrote, a software you’ve tried to use unsuccessfully, etc. It is tempting to move on as soon as possible after wasting time on something that didn’t work, but do save your future self the time by making notes about what you’ve tried already.
   • Notes about every meeting with your supervisor - is the project progressing well in your supervisor’s opinion, what deadlines you are working towards, what should be your priorities, etc. Share these notes with your supervisor through an email or a shared folder. If there is ever a conflict, these notes will be invaluable evidence in your defence.

3. Keeping your files organised. You’ll have to find a system that works for you. In general, it’s a good idea to keep data separate from your analysis of the data, to name folders with a date and a topic, to keep copies of software you’ve used in case it gets updated and you can’t access an older version anymore. See this paper (even if you are not a computational biologist) for a great list of suggestions, try things out and don’t give up until your files are so well organised that you never lose anything.

4. Following literature using RSS feeds. A lot of junior scientists, even post-docs, rely on their supervisors to notify them about important papers coming out in their field. Often PhD students discover a relevant paper, which they should have been aware of, only when writing the literature review section of their thesis. To become an independent and well informed scientist, I highly recommend following relevant journals through their RSS feeds using an RSS reader app (Feedly is my favourite). First, it won’t take any extra time as you can scroll through the app while waiting for a bus or for a kettle to boil. Second, by scrolling through all recent publications (instead of relying on keyword searches) you’ll become aware of developments in fields adjacent to yours, and become more confident at networking events. Your supervisor will be amazed when you show them relevant papers that even they are not aware of. If you’ve never used RSS feeds before, it might seem unintuitive at first, but spending time setting it up is totally worth it. If you don’t know where to start, add Nature RSS feed first. Apart from great science in a diverse set of disciplines, they also publish articles about PhD student mental health, tips for networking and scientific presentations, etc. Once you find papers relevant to your work, add those journals to your reader app.

   Once you are using an RSS reader app, you can also use it for personal benefit. Add websites you typically visit (like Wired or Visual Capitalist); when you come across non-scientific articles you’d like to read in your spare time, send them to another app (Pocket is my favourite and it works offline). I find that my brain archives languages I’ve studied for a bit but then stopped studying/using; I struggled with re-learning Spanish many times. The fix I discovered is adding an RSS feed in Spanish to my reader app. Even if I never actually read any articles in Spanish, just seeing a couple of titles in Spanish every day sends a signal to my brain that the language is in active use and I keep learning some new words.

5. Getting access to your university’s VPN and Eduroam. The VPN (if available through your university) might not be a great one, and it might be a hassle to get it working, but it will allow you to freely access pay-walled papers when working from a different location. Often you can keep your access to the VPN for much longer than you get to keep your student card, so it will be handy even when you graduate. Another invaluable thing is Eduroam - it’s a WiFi which you can use if you have a username and a password for it (sometimes it doesn’t work immediately and some setting changes on your devices are required). There might be a different WiFi in your building or you might be on a wired connection, so it might seem irrelevant, but you absolutely must make sure that you can connect to Eduroam (if available through your university) on all of your devices. That is because the same username and password will work on all Eduroam networks, including many universities in 106 different countries and even Geneva Airport. Once connected to an Eduroam network of any university you’ll be able to access papers from pay-walled journals to which this university subscribes, even if your university has no access to them.

Doing a PhD

These are the five skills I found most valuable while doing a PhD:

1. Adjusting your mental zoom level. Often you’ll be faced with a task so big that you’ll just sit and stare at it without making any progress for days. In that case, instead of hiding behind a mountain of relevant papers that you absolutely have to read, you should zoom in on a particular doable sub-sub-problem of a sub-problem of a problem you are working on. Find something that can be done in less than a day and enjoy the feeling of having made some progress. This blog post describes the value of focussing on making progress on the problem instead of on solving the problem.

   Often you’ll feel like you’ve been doing something for weeks and you still feel stuck. At that point you should zoom out and remind yourself what is the big problem you are working on. Are the details you are concerned with really important? Did you have in mind a couple of ways to approach the problem and now you are stuck on one of them? What achievable goals do you have to complete to be able to move on? Ultimately, no matter what’s happening in your field or with your project, you have to submit your thesis, graduate and build your path to the future you want to have. Find a way to get yourself unstuck by looking at the big picture, even if it’s not a pretty picture, and making a concrete plan.

   An advice you’ll probably not hear from anyone else - read “The Martian” book. It wasn’t intended that way, but the book touches on many aspects of a PhD life - longing for guidance and support from your supervisor, longing for your supervisor to stop telling you what to do, feeling lonely, surmounting despair, bad situations getting worse, staying organised and creative, and finally a happy ending.

2. Navigating the ocean of papers. It will take time to learn to distinguish good papers from bad papers. An obvious one is if you see the words “we hope” in the methods section, but most of the time it will take more skill to judge the paper correctly. It doesn’t help that often bad papers are cited more frequently than good ones and hence are easier to find. A good place to start is from a good review paper of your field, from there you can go both backwards (the papers they cite) and forwards in time (the papers that cite this paper, which you can find through Google Scholar “cited by” feature).

   At some point you might figure something out that is contrary to what everyone else in your field believes. Surely it can’t be that everyone else in the whole world is wrong? Do subject your point of view to a healthy level of doubt, but don’t discredit yourself - it can happen that everyone else is wrong, as this recent example shows.

3. Mastering social interactions. Every academic field has a unique dynamic of how social interactions and collaborations work on the level of a research lab, a department and a whole field across different universities. If you are very lucky, it might resemble the dynamics described here, but in most cases you’ll be surprised how counterproductive it is. Learning to work around the dynamics of your field and building lasting personal connections with people in your field (even if your supervisor doesn’t talk to that person) is essential for making a contribution to your field. Equally important is to learn to discuss your work (in detail) with people from other disciplines. This specific example is probably irrelevant to you, but there is a lot to be gained from reading this paper.

4. Producing science. It is irrelevant what you do, only what others can find and build on is actual science. Writing papers and your thesis, and finding ways to make your data and your code publicly accessible is important. For me the writing part was the most difficult one - it was not unenjoyable and I was rarely happy with the text I produced. I spent a lot of time staring at an empty file, struggling to write the first sentence, then rewriting it, struggling with the second sentence, and then going back to rewrite the first sentence again. What really helped is starting with a skeleton and then slowly adding some meat to it. Even if you have no idea where to start with writing a paper, you know you’ll probably have an introduction, a methods section, a results section and some sort of a conclusion. That’s a start! From there you can make a bullet point list for each of these sections, keeping points in a random order for now. Once you can’t think of any more points to include, you can then arrange the points in each section in some sort of logical order. There are many different ways to arrange several points that are all interconnected into a linear structure of written text, and often none of them is clearly best. Just pick one and get over it. Once all of your points are arranged, focus on one point at a time and decide roughly how many paragraphs you’ll write for each of them and the topic of each of those paragraphs. Then you can make a list of points covered by each sentence of a paragraph. Before you know it, the paper is pretty much written even before you’ve started doing any actual writing.

   Look through several guides for good academic writing (this one is a good example) and find the one that fits your personal style. Reading a lot of papers and focussing on the ones I found enjoyable to read really helped me to improve my writing. Some papers are difficult to read and you keep finding yourself losing track and thinking about something else, other papers are a delight - you read them quickly, evocative analogies make vivid images in your brain, jokes make you laugh and you start feeling like you know the author of the paper personally. You definitely want your papers to be more like the latter ones!

   When you are a young scientist it is tempting to prove you are not an undergrad anymore by using overly complicated sentences and writing long paragraphs peppered with highly specialised words and as many acronyms as possible. But your contribution to science will be so much greater if you make your writing easy to consume, which means more people will actually read it. When proofreading my own writing, I focus on the question “What will the reader learn from this particular sentence and this particular word?” and ruthlessly delete everything that can be deleted. Don’t ever worry about your paper or your thesis being too short. Everyone is busy, and your skill to communicate ideas and facts with a minimal number of words is valuable.

   Don’t hesitate to try out new ideas. For example, to help your reader visualise a process inside a biological cell or a dynamic of celestial objects consider scaling everything to sizes of objects typically seen in a room. Say a skin cell is about the size of a sofa, the compartment of the cell you are studying is the size of a cat and the enzyme that you’ve discovered (as big as a poppy seed) interacts with the protein the size of a hazelnut. You should still include your measurements expressed in microns, but it’s the vivid image you’ve created with objects the reader is familiar with that will allow them to easily imagine what you are talking about and stay with you for the rest of the paper. I’m sure there are loads of creative ideas like this one out there, it’s a shame they are rarely seen in actual papers.

   One thing you should be really careful with is choosing colour schemes for your plots, see a good explanation here. Often conclusions made from examining the plot do not hold if the colour scheme is changed. This is due to how our eyes work and how some pairs of colours are perceived as more similar than other pairs even if the objective distance between the colours in the colour space is the same. There is no “best” colour scheme, so remember to look at your plot in several different ones before making your conclusions. It is a good idea to make your figures color blind people friendly.

5. Making the most of your time as a PhD student. Being a PhD student is a great platform from which you can shape your future. If you want to develop a particular skill (data analysis, web-design or organising events) you can always find a small project or an interesting collaboration that will allow you to do so. If you want to build connections in a particular country, perhaps to move there for work after graduation, you can travel there for conferences or even arrange a collaboration which will allow you to be a visiting student in one of the universities there for a short time. Do you want to speak with an expert in the industry or arrange a visit to a factory? You’ll be surprised how many doors will open if you simply send an email to the right person saying “Hello! I’m a PhD student from X university and I’d like to…”. It doesn’t even have to be in a relevant field, PhD students are allowed to be curious. A similar email saying “I’m a recent graduate” (which sounds like “I’m desperate for a job”) might not open the same doors. Do you have an idea about which companies you might want to work for after graduating? Get to know people who work there while you are still doing your PhD. Your supervisor might say you don’t have time for all this, but please don’t listen - there is nothing more important than your future and there is no better time to start building it than today.

Surviving your viva

Nobody will ever be as interested in the work you’ve been doing for the past several years as during your viva. Get into the mindset that it will be an enjoyable discussion and that you have as much right to steer it as your examiners do. That said, the goal is to have as little corrections to do as possible. That is because almost nobody will read your thesis, so working on your papers and taking next steps in your career is much more important both for you and for science in general. Here are my top five phrases to use during your viva to ensure that you get minimal amount of corrections:

• “This is beyond the scope of the project.” - if the examiners say you forgot to do something, emphasize that in the limited time you had you executed a well designed project, this just wasn’t part of it.

• “You are right. I think it’s best if I remove it from the thesis all together.” - if the examiners say that something was done to insufficient depth or requires more validation. If it’s a small thing, don’t agree to do any corrections related to it! Emphasize that it was tangential to the core project and it’s your mistake for including it into the thesis in the first place. To have a thesis with a good story, it’s best to simply remove this tangential part.

• “No, this was not the aim of this project. The aim was…” - if the examiners say you haven’t fully achieved a goal of a project, don’t hesitate to correct them. Clarify that the examiners misunderstood your goal, improvise a good story about what the actual goal was and do not agree to the proposed corrections. If you know which parts the examiners liked, emphasise that exactly those things were the main goal.

• “I agree, it would be great to do this! Unfortunately to do it well, it would take a whole new PhD to accomplish. I don’t see much value in doing it badly.” - if the examiners suggest some really big corrections, this is the best way to get out of the looming danger.

• “Absolutely! I’ve just added this to my list. This is a brilliant suggestion. This will really improve my thesis, thank you so much for this suggestion!” - if the examiners give you a correction that takes 10min to do, agree immediately and loudly. Overemphasise it as much as possible, to make sure the examiners feel like your list of corrections is filling up and that they should not give you too many more.

Making the most of your new title

After doing a PhD you have more options open to you than before you started your PhD. If you haven’t even started your PhD yet, this will seem obvious to you. Surprisingly, this might no longer be obvious to you after you graduate. Well-meaning people around you will tell you that you now absolutely have to pursue a career in the area of your PhD, and that all the other directions are now essentially off limits to you. Some will even say that if you don’t pursue a career in your field then you’ve wasted years of your life by doing a PhD. Just remember yourself before you started your PhD and you’ll know this is not true. You’ve learnt a lot of soft and technical skills, you’ve learnt how to learn quickly and how to manage years long projects, you’ve learnt to work independently and question everything. Your future will be as amazing as you’ll make it. There is nothing you have to do, you are free to go and chase your dreams.

If you’d like to discuss anything mentioned here, or if you think I can help you to make the most of your PhD, please do get in touch.