Work

I do cancer genomics. The through-line is measurement: taking properties of tumours that are agreed to matter — how unstable a genome is, which route it took to its present state — and turning them into something a sample can actually be scored on.

That means methods as often as findings, and software as often as papers.

Cancer Genomics

What tumour genomes record about the processes that shaped them, read off sequencing rather than inferred from models alone.

Chromosomal Instability

CIN exists at levels that matter clinically, but the rate itself is hard to measure. Most of my work has been on quantifying it.

Complex Tumor Evolution

Genome doubling and multipolar division push tumours into copy-number states that no simple sequence of events explains.

Phenotype Plasticity

How much of what a tumour does is written in its genome, and how much is a state it can move into and back out of.

Methods & Software

Tools that make the above measurable: inference of mis-segregation rates, and figures built to argue rather than to browse.

Writing

all posts →

Publications

all publications →

Software

Project What it does Source
CINFER scDNAseq-based estimates of chromosome mis-segregation rates. Repository ↗
ggsuperviolin Violin SuperPlots for ggplot2 Repository ↗
iBover Models the karyotypic consequences of multipolar mitotic divisions.
SeqPlotR Flexible, grid-based genomic track plots. Repository ↗

everything else on github →