21/01/2021
A masters student emailed asking for advice with their PhD application, this is what I said (with some changes) - you may find this helpful, if not: that’s OK. This is just my experience and what I thought about the process.
Note: my application was in bioinformatics, this advice will differ for other topics however could be useful still.
Skills to highlight in the application essay?
I would say start with your masters (MSc), what skills did you recently learn and how? Did you learn about genetics via a module or a programming language via a group project? For example, I did statistics - and despite not doing so well in the statistics module, it’s important I mentioned it because I CAN do it.
Talk also specifically about the skills you acquired during the MSc dissertation: what your topic was and what skills you used throughout. Personally, I handled nanopore sequences and used various software to explore the data, such as a BLAST to observe species or looking at gene annotations with the Gene Ontology, but also I made my own Python package to answer the main research question.
Also talk about potential skills - maybe a computational technique you know of but perhaps you haven't used yet. I didn't use machine learning extensively in my Masters however I know of it from one of my modules with practical sessions. I remember some other brief skills I highlighted included some from my undergraduate, since everything is relevant in my academic career, only I highlighted the ones which were important for the PhD. I also mentioned generic skills, “presenting”, “enthusiastic”, and “independant”.
It does depend on the type of PhD you are aiming to do/apply for, for example if you were applying for a Genomics-based PhD then perhaps mention more specific skills/knowledge: genes, maybe you can run a GWAS, or the statistics that are frequently involved (e.g. with a GWAS you would use logistic regression) in that subject area. I remember in my application, I wrote a paragraph about how I interpreted the research question and what I thought about possible methods.
What was the interview procedure like?
I actually remember my PhD interview very well! I walked in, introduced myself, and shook everyone’s hand. There was my supervisor, someone from HR, and staff who were a part of my funding body - however some people may have only a supervisor and a postdoc present.
I have heard it’s common that some panels require you to present something. My interview required that I discuss a research topic I recently did, “my best piece of research”, so I talked about my MSc dissertation. Some people may not have to do this! After the introductions, I led straight into a presentation. Afterwards, they asked some questions about my presentation, some regarding skills/experience, and finally a few questions about me.
I felt my interview was fairly relaxed, however I do understand not everyone has the same experience. My supervisor was quiet for most and it was a little intimidating, but it gave them the chance to listen and asked me the technical questions. The others on the interview panel majority asked questions about me, to get to know me. I believe I was able to express my passion for research.
What did I present and what skills did I mention?
My presentation was my MSc dissertation: the topic and skills. Only a few skills I learnt/used were specifically relevant for the PhD application: but it was still important to mention everything. My presentation topic was: “My best piece of research and why this?” I started with a little introduction about myself, my name, and that I studied my Masters in Data Science. I decided to present my MSc dissertation because I really enjoyed it and I learnt a lot. I presented my work as if it were a PhD, I personally thought it shows a research project and had a nice flow.
GC content of the sequences from the phylum seem to be consistent in groups, after a brief observation of sequences publicly available of the phylum this seems to be true" - I presented that I could do background research.
GC content?" - a clear and defined research question based on my introduction and literature.
Python package, which groups sequences based on current available data" - technical skills should be mentioned here: I can code in Python, I can obtain data, and I can sort/classify data.
GC content: this could be useful for future knowledge of the phylum" - I contributed something to science!
Are there other possible presentation topics?
Some panels require different topics to be presented and give you different times, some only 7 minutes then 3 minutes direct presentation questions. Others state a 10 minute presentation (then ask you about it afterwards). You may even be asked to present for 15 minutes, or 3 minutes with 2 minutes of direct presentation questions.
What sort of interview questions will come up?
After my presentation, the first few questions were about the presentation. One for example: "what would you do differently?" I always find this one difficult because it can be hard to critique your own work, especially since I didn’t know much about other methods back then. However I remembered from my dissertation some things I wrote in the conclusion, and so my answer was something like, "I think in future I would look into some machine learning prediction analysis, as I do have the skill from the masters course".
After the initial presentation questions, the questions turned more technical about my knowledge and experience. One question I remember clearly from the supervisor, "what would you do if I gave you a dataset of patients with heart disease?". Despite doing a MSc module on machine learning, I felt like I knew so little about it, but I remember clearly that I mentioned how data cleaning is the first step and then I threw in some buzzwords like, "supervised" and "unsupervised" methods - they liked my response, but wanted to know more, and asked what I meant more specifically. I vaguely remember saying something like "Using Python I would do unsupervised clustering or possibly a supervised random forest decision tree. But it’s tricky because it depends what sort of data it is and how to fine tune the methods." This response shows I know about Python (or R if you use that), and some basics of machine learning via using some buzzwords: that I know about unsupervised clustering methods.
I found in my interview that the panel wanted to know things about me. It felt like they were trying to see if I was a good fit for the PhD and the funding body. For example, my particular funding group of students have planned activities: I think they wanted to see I would be the type of person who would be involved in these. They asked questions like, "how would someone describe you?" and the question everyone dreads, "what is your biggest weakness?". I mentioned my biggest flaw was being shy, and despite this I was working hard to overcome it. They appreciated the honesty and agreed that overcoming shyness was difficult but necessary for the PhD activities (linking back to these funding activities).
I mentioned I was described as a friendly person, hard-working, etc. I’m not sure what the panel thought about these generic responses: but they were the truth. I couldn’t properly describe myself as a researcher yet as I didn’t have that full experience: I was not aware of the type of PhD student I could be but all I could answer were my honest opinions about myself.
There were other questions like, "do you enjoy research?" or "what do you think about the PhD project?" and I think they appreciated my enthusiasm! I love research personally because it gives me the chance to explore data using my skills but also learn stuff along the way. I mentioned I love presenting my work - they seemed to love that response, probably because supervisors love having their students show off their work via presentations / seminars.
Are there any skills or general bioinformatics skills to highlight?
Skills might be mentioned in your presentation (if you have one), others might come up in the questions, you'll probably mention skills in the essay. I think the panel will ask about skills which they think are relevant, e.g. the supervisor asking about how I’d handle a dataset and asking me to expand on machine learning. I would definitely mention the skills that are popular in the area, e.g. a skill in bioinformatics is being able to code. Also statistics is a good skill to mention, but don’t worry if you don’t mention it: I didn’t mention it at all (only in the application essay). I feel like it’s just a few buzzwords you need to know of your skills, e.g. currently I would say my skills in terms in buzzwords are: unsupervised machine learning: clustering; statistics: dimension reduction; natural language processing: word embeddings. But if I were to have an interview today, I would mention these in terms of projects/papers: what skills I used during particular projects to reflect for the job I'm applying for.
Some possible general skills to mention: presentation and communication skills; good listener and a keen learner; enthusiastic, independent student. During my PhD, I have found that I’ve become very independent and I'm still enthusiastic about my work. My communication skills have increased and I’m becoming more confident when I talk about my work during meetings or seminars.
Are there any trick questions to look out for?
This is a tricky one: I was prepared for my interview, except for one thing: I didn’t do much background research into the supervisor! A question that was asked, "what do you know of the supervisor?" This one surprised me, but I knew of the supervisor as they were previously at my university and staff members spoke fondly of him. However I didn’t actually look into their research or papers, I knew they did ontologies, which I too enjoyed, so I mentioned that. But I’m not 100% sure if they were hoping for more.
The only trick question that I definitely didn't expect at all: "out of all your supervisors' papers, which one did you like the most?" - this one caught me off guard! Some people I spoke to afterwards said this was a very common question, so I wish I knew about it! I was completely honest and said something like, "I am really sorry but I haven't read any of their papers. But I have heard of *supervisor* and people have spoken fondly of them". I think they appreciated the honesty, and my supervisor seemed happy.
Any final tips?
The PhD is a time to learn about various methods and techniques, so one of my tips would be don’t worry if you can’t answer a question, or even at all. As long as you have some response, if it’s “...sorry but I haven't read any of their papers…”, then they will appreciate it. But again, this is my opinion.
I personally believe a good tip is to be yourself: I felt like the main part of my interview was that they wanted to see what I wrote for the application was true. They wanted to see that I knew about the skills I put down on the application - there’s no need to lie, or exaggerate the truth. The panel wanted to know who I was as a potential researcher, they’re not there to be mean or trick you, rather to catch out people who may have written down stuff on the application which wasn’t true, yet was relevant for the PhD.
I put keywords in bold throughout my presentation slides, e.g. "sequences, Python, machine learning, bioinformatics tools, ontology". To highlight specific skills and knowledge. This really catches the eye! Furthermore I tried to say some buzzwords, “data cleaning” and “unsupervised”. Even if I never use particular skills in the PhD, by using buzzwords in my application and interview they knew what I was capable of and future potential. Some other terms that could be useful, "statistics", "GWAS", "sequencing", “genes”, “traits”, “phenotype”, and more.
Finally, maybe do a little background research of your supervisor, even it’s a few key-words, e.g. you can playback to the panel words like: ontology. But of course make sure you know what these things are! You don't need to know the inner details and the algorithms, as long as you know something about these buzzwords, your skills, and yourself: I think you'll be fine.
Good luck!
The PhD is a learning and development journey, so even if you don't fully understand some things try not to worry: the PhD is a time for the supervisor to pass on their knowledge and you’ll learn other stuff. Good luck to anyone reading this who is preparing for a PhD application/interview. It’s not a personality contest! They just want to know you: if your skills and knowledge are fit. It’s OK to not know everything related to the PhD proposal, the PhD is a time to learn along the way, it’s OK if they ask a question you don’t know the answer to. Who knows what the panel think! I walked in my interview nervous and came out feeling positive! These are all my opinions and experience - everyone is different - and the panel will ask different things. Despite feeling like the panel wanted to know more about me and my skills, I don't actually know what they thought or said afterwards.