Medicinal chemistry · Computational biology · Machine learning
Understanding biology through computation.
From molecular interactions to multi-omics discovery.
I'm a medicinal chemistry graduate building computational research at the point where chemistry, genomics and statistical learning meet. This site holds the work itself: published analyses with their data and limitations, live tools you can use in the browser, and the research questions I want to pursue next.
- 12 published analyses, each with stated limitations
- 10 interactive tools built on real scientific data
- Every result traceable to its source
Open to doctoral research opportunities and computational biology positions
Live workbench
Different readers need different evidence.
A doctoral supervisor, a hiring manager and a collaborator are looking for different things. Each route below is curated for one of them, drawing on the same underlying work.
-
1
Research
Scientific direction
Research interests across five connected themes, three independent research proposals, and the reasoning behind the methods I choose.
- Research themes 5
- Independent proposals 13
- Methods library 34
-
2
Engineering
Technical work
Published analyses with code, interactive tools built on live scientific APIs, and two production systems serving real users.
- Published analyses 12
- Interactive lab tools 10
- Production systems 2
-
3
Professional
Background and CV
Education, the medicinal chemistry dissertation, technical capabilities with evidence attached, and how to get in touch.
- BSc Medicinal Chemistry Cardiff
- Dissertation case study 40 credits
- CV Academic + industry
Two disciplines, deliberately.
A computer scientist can build a machine learning pipeline without the chemical intuition to judge what its output means. A medicinal chemist can reason about molecular interactions without the engineering to work at the scale biological data now demands.
- Chemistry first. Three years of synthesis, characterisation and molecular design, including a 40-credit research dissertation involving catalysis, analytical characterisation and docking.
- Computation second, and deliberately. Genomic annotation, population genetics, statistical genetics and machine learning, built project by project against real public APIs.
- Engineering throughout. Two production systems with databases, sandboxed execution, automated testing and continuous deployment.
- Honesty as a method. Every project states what its data can and cannot support. Several exist specifically to document where a method breaks.
The work, by theme
All projectsHover a tile to see the theme. Click to filter the project directory.
Status, honestly
Start here
MoreEvery capability, attached to evidence.
Percentage bars mean nothing. Each bubble below links to the work where that capability is actually used. Tap one to see where.
Only technologies used in work shown on this site are listed.
Four projects worth ten minutes.
The first is a finding; the rest document something that went wrong, or something a standard method gets wrong. That is usually where the useful part is.
Pick the route that fits.
PhD pathway
Research readiness: scientific questions, methodological reasoning, statistical understanding and independent proposals. With specific notes for UCL and the University of Sydney.
- Research proposals 13
- Method notes Open
- Academic CV Download
Industry pathway
What I can build: genomic analysis, predictive modelling, molecular work and the software engineering to make any of it reproducible by someone else.
- Production systems 2
- Capability matrix 11 areas
- Industry CV Download