Have you wondered what a Biostatistician does? Andrew Grant has given a snapshot into what a typical day for him involves.
Andrew is a Research Associate at the BSU. He works on developing statistical methods to find out possible risk factors for diseases.
What did you do before joining the BSU?
I completed my PhD in Statistics at Macquarie University in Australia in 2018. I then worked there as an associate lecturer in the department of Mathematics and Statistics for 6 months before joining the BSU.
What does a typical day in your role involve?
Daily activities related to my research are varied, depending on where I am at with a research project. Working through research problems can involve a lot of reading, trying out new approaches both ‘on paper’ and by coding on the computer, and then writing up results and developing software to go along with them. Interspersed with these are meetings with my research group and other collaborators, as well as the wider department. There are also a host of seminars that I am able to attend, such as those regularly hosted by the BSU. One of the positive aspects of lockdown is the number of online seminar series that have been established, meaning you can ‘attend’ talks with other researchers all around the world.
What is your work schedule?
I typically keep to business hours for working. However, if ideas pop into my head during the evening or weekend, I usually try to think these through at the time before I forget them!
What does your work space look like?
At the office, my desk has a monitor, keyboard and mouse, and is usually surrounded by many loose pieces of paper with scribbled notes and a few textbooks. I try to keep this all organised but I’m not always very good at this! During lockdown, I’ve made a similar setup in a spare corner of my house. Thanks to the excellent IT and admin support we have at the BSU, I can work effectively pretty much anywhere I have my laptop and WiFi connection, which has come in handy this past year.
What are you working on at the moment?
Broadly speaking, my research centres around ways to distinguish between ‘correlation’ and ‘causation’: we often observe variables in a dataset which show a pattern of association, but this doesn’t necessarily mean that changes in one variable will cause a change in the other. Making this distinction is important but often difficult because of unknown factors which might mask the true relationship between the variables. More specifically, I work on methods development for Mendelian randomization, which is the use of genetic variation to study whether potential risk factors (for example, increased BMI or blood pressure) cause increased risk of disease (for example coronary heart disease). Mendelian randomization can be a very powerful tool but relies on a number of assumptions to be made. I am particularly interested in developing methods for performing valid Mendelian randomization analyses when these assumptions are not justified, and also when other obstacles are faced such as when full datasets are not available for analysis but only summary statistics from these datasets.
What is your favourite part of your job?
It’s very hard to pick just one! Mostly it’s the fact that I get to do things that really excite and interest me every day and can call it a job! Being able to work with the vastly experienced and talented people at the BSU, and the opportunity to collaborate with people in other fields, particularly with applied and clinical researchers, is a major plus.
What is your least favourite part of your job?
I’m not sure that there is a part I don’t like – some aspects can be challenging, such as trying to learn topics in new areas, but these can also be the most rewarding aspects when the challenges are overcome.
What advice would you give to anyone considering your role?
I think anyone who enjoys working through challenging statistical problems and has an interest in applying them to important real world settings would find the job very rewarding.
A bonus is that Cambridge is a really great place to live and work.
Blog post by Andrew Grant