For doctoral supervisors and admissions committees

Research readiness, shown rather than asserted.

The case I would make is not that I already know how to do a PhD. It is that I can identify a question, choose a method for a reason I can defend, implement it, and recognise when the result does not support what I wanted it to say. This page gathers the evidence for that, including where it is thin.

Read the proposals Research programme Methods library Email me

Seeking doctoral research in computational biology, bioinformatics or precision medicine

What to look at, and why

Four pieces of evidence.

Ordered by what I think is most informative about research potential, rather than by what is most impressive.

Research thinking

The habits I'd bring to a group.

These are visible in the published work, which is the only reason I am willing to claim them.

  • I report the result that undermines the method. The VEP project exists largely because SIFT calls a well-established functional variant "tolerated". That is the interesting part, and it is in the findings rather than buried.
  • I check against a case whose answer I know. The gene annotation pipeline silently returned "no overlapping gene" for a variant that is textbook MTHFR. It was only caught because the right answer was known in advance.
  • I separate the layers of a claim. Transcript annotation, allele function and clinical guidance are three different kinds of evidence. Collapsing them is how pharmacogenomic overclaims happen.
  • I state what the data cannot support. A fold-change analysis with one value per condition gets no p-value, and the project says so rather than quietly implying significance.

Where I am weakest

I would rather you hear this from me than discover it in an interview.

No wet-lab biology experienceMy laboratory work is chemistry, not molecular biology Statistics are self-taughtSolid on fundamentals; no formal graduate training No large-scale omics analysis yetMy published work uses small datasets; the flagship is proposed, not done No publicationsIndependent outputs only, not peer reviewed

A 2:1 also sits below some programmes' stated entry requirements, and I am aware that is a real constraint rather than something a portfolio overrides.

Institution interest

University College London

Applying · 2027 entry

Mapping spatial plasticity in colorectal cancer

A PhD at the UCL Cancer Institute asking whether every region of a tumour is equally able to adapt to treatment — multi-region organoids, mass cytometry and spatial transcriptomics, linked to the REWIRE-CAN Cancer Grand Challenges team.

I have written a full proposal in response: hypothesis, the experiment that separates intrinsic from environment-induced plasticity, how I would define a plasticity index, and the five ways it could fail. Read it →

What I would prioritise showing

An honest note. I have deliberately not written a paragraph claiming deep familiarity with specific UCL groups. Before applying I would read a prospective supervisor's recent work properly and write to them about it specifically — which is a different exercise from a portfolio page, and one that should not be faked here. No affiliation with UCL is held or implied.
Institution interest

University of Sydney

Why Sydney

Sydney's interdisciplinary research culture, and the Charles Perkins Centre's model of putting computational, clinical and biological researchers on the same problem, fits the way I want to work — at the join between molecular detail and systems-level questions rather than inside one discipline.

The proposals most relevant here: pharmacogenomics, computational drug discovery, and network-based target investigation.

What I would prioritise showing

Same caveat. This is a statement of my interest, not a claim of any connection to the University of Sydney. Specific supervisor alignment is something I would research properly and discuss directly rather than assert on a web page.