Curriculum Vitae

I worked as a biochemist, both in academia and industry, for about nine years. During my PhD, I started using code to analyse my data, and using machine learning to find patterns1. As well as continuing my development as a wet-lab scientist, I also improved my skills in machine learning and bioinformatics. I was really excited to have the opportunity to put my own stamp on each of my academic projects, including bringing image analysis and deep learning into my experiments2. My time working in a pharmaceutical contract research organisation gave me a new appreciation for the challenges of working with larger datasets.

Recently, I have started working full-time3 using my data science skills. Most of my time is spent developing a tool that uses deep learning to harmonise health data, though I’m involved in a few other data projects. I’m really enjoying using that part of my brain more, though it’s very strange not using a pipette.

Professional History

University of Nottingham

Postdoctoral Research Fellow July 2024 -

I work on a tool, LLettuce, that applies deep learning to healthcare data. The purpose of this is to help data engineers map their data onto standardised medical vocabularies so that it can be compared with other data without sharing confidential information. Most of my time is working for the School of Medicine, but some is for the Digital Research Service, so I’m gaining valuable software development experience, too.

Excellerate Biosciences

Senior Scientist October 2022 - April 2023

I ran in vitro screening assays for clients and developed new assays. As well as being responsible for routine analysis of screening data, I also carried out bespoke analyses as required, often preparing material for client meetings.

University of Nottingham

Postdoctoral Research Fellow February 2019 - September 2022

I worked as a postdoc on a project investigating the transport pathway of the multidrug transporter, ABCG2. I was lucky that the investigators on the grant were open to new ideas, so I developed novel assays and novel bioinformatics methods as part of my work. This included bringing supervised and unsupervised machine learning into the analysis of biological data. I also really enjoyed being able to make GUIs to help colleagues without coding experience use tools, and develop interactive dashboards to communicate results. I presented work from this project at national and international conferences, and some has been published4.

University of Warwick

PhD, Medical Sciences September 2014 - April 2018
BSc, Biochemistry September 2011 - June 2014

My PhD project developed rhodopsin as an experimental system for membrane protein folding. Part of this involved characterising misfolding mutants, and a detour into characterising the interactions between rhodopsin and a small molecule. I was also part of a really fruitful collaboration with scientists at the University of Cambridge. This all gave me a solid grounding in the application of biophysical methods, as well as my first taste of how satisfying it can be to carry out the right analysis for your data.

Tools

Python

I’m most comfortable carrying out tasks in python, with plenty of experience with the usual suspects in a data toolkit5. If you look in my projects, you’ll see some more unusual libraries being used, which are just the tip of an iceberg of tools I’ve picked up once or twice.

Javascript

I found javascript quite hard to begin with, and some parts of it still seem pretty weird to me. However, I really like the functional features that have been brought in, and if you want to make something interactive to share with people, it can’t be beaten!

R

My first forays into code were with R, simply because the one biologist I knew well who could code used it6. I miss it sometimes, and the tidyverse is a lovely bit of joined-up design for handling data.

Microsoft stuff

It’s hard to work anywhere people do stuff with numbers without encountering Excel. I have always been pretty comfortable with it, and in my last job I got a lot more comfortable with its more advanced features. This included writing a fair bit of VBA, which I will do if you ask very nicely. In the time since, I dug a bit deeper into Microsoft’s offerings, and I’m capable with Power BI and Power Query.

SQL

I had read online about how SQL and relational databases were old fashioned and going to be obsolete. Then I read a bit, and designed a database, and the ideas behind relational databases like normalisation just felt very satisfying. I’m no SQL wizard, but I can create a useful data model and write a moderately complex query.

Footnotes

  1. At the time, I kept finding myself reading about machine learning and thinking “isn’t this just stats?”↩︎

  2. papers in preparation↩︎

  3. OK, four days a week↩︎

  4. I should probably be writing the rest right now!↩︎

  5. pandas, numpy, scikit-learn, matplotlib↩︎

  6. Hi, Joan!↩︎