Dharmesh D Bhuva
Portrait of Dharmesh D Bhuva

Dharmesh D Bhuva, PhD

Computational Systems Biologist · NHMRC Emerging Leadership Fellow

Frazer Institute, The University of Queensland · Honorary Research Fellow, South Australian Immunogenomics Cancer Institute (SAiGENCI), The University of Adelaide

About

I am an early-career, NHMRC-funded computational systems biologist on a mission to understand how complex systems of gene regulation and signalling produce diverse tissue phenotypes in health, disease, and development. My current work develops computational methods and frameworks that generate accurate biological insights from spatial molecular datasets — from high-plex, sub-cellular resolution spatial multiomics through to whole-transcriptome tissue atlases.

I completed my PhD in October 2020 through the School of Mathematics and Statistics at the University of Melbourne, developing novel systems biology approaches to study molecular function and gene regulation in cancer. Since then I have worked across WEHI, SAiGENCI, and now the Frazer Institute, where I lead work on spatial biomarkers and topological analysis of cancer tissue and its microenvironment.

I build open-source software used worldwide — including singscore, SpaNorm, standR, and vissE — and I am an active teacher, workshop organiser, and supervisor of graduate research students.

24
Publications & preprints — 10 first-author, incl. 4 in Genome Biology & 4 in Nucleic Acids Research
13
Open-source R & Python packages and web apps
272k
Total package downloads, updated on every deploy
$3.7M+
Research funding (NHMRC, MRFF, ALCCRF & more)

Current focus: modelling counts in spatial transcriptomics

Spatial transcriptomics measures gene expression while preserving where in the tissue each measurement came from, and that context changes what the measurement is. Effects safely treated as nuisance in dissociated data — library size, the identity of neighbouring cells — become spatially structured and entangled with the biology. My work builds count models that take this seriously.

A cell's behaviour is not a property of the cell alone. The same cell type can respond to a treatment one way when surrounded by immune cells and another way at the edge of a tumour nest, and a conventional differential expression test — which asks only whether a gene changes with the condition — averages those opposite responses into nothing. spiDE asks the sharper question: does the condition effect itself depend on the local neighbourhood?

Multi-scale niches kernel density of each cell type, measured at several bandwidths covariates Negative binomial GLM per gene, on SpaNorm's fitting engine log μ ~ cell type × condition × niche does the condition effect depend on the niche? Combine & control Wald tests, combined across covariates (Brown) + bandwidths (Cauchy) Hierarchical FDR gene → index cell type → niche cell type Context-specific DE genes
spiDE summarises each cell's neighbourhood as kernel density estimates of every surrounding cell type, computed at several bandwidths so that effects at different spatial scales can be detected. These niche covariates enter a per-gene negative binomial GLM — built on SpaNorm's fitting engine — containing a cell type by condition by niche interaction. Because the resulting tests are heavily correlated across covariates and bandwidths, evidence is pooled with Brown's method and Cauchy combination, and error is controlled hierarchically from gene down to niche cell type.

spiDE is built on the negative binomial fitting engine from SpaNorm, my spatially-aware normalisation method (Genome Biology, 2025).
More on my research — normalising counts, neighbourhood-dependent differential expression, and multiplex proteomics →

Research interests

  • Spatial transcriptomics & multiomics
  • Cancer systems biology
  • Gene regulation & signalling
  • Statistical bioinformatics
  • Tumour microenvironment
  • Open-source scientific software

Software highlights

singscore hex sticker

singscore

Bioconductor · since 2018 · 144,605 downloads (top 10% globally)

Rank-based, single-sample gene set scoring to quantify molecular phenotypes from transcriptomic profiles of individual samples.

SpaNorm hex sticker

SpaNorm

Bioconductor · since 2024 · Genome Biology 2025

Spatially-aware normalisation that removes library size effects from spatial transcriptomics data while preserving the underlying biology.

vissE hex sticker

vissE

Bioconductor & vissE.cloud · since 2021

Identify and visualise higher-order molecular phenotypes from gene set enrichment analyses; also available as a code-free web server.

See all 13 packages and web apps →

Appointments

  • 2024 – Present
    NHMRC EL1 Investigator
    Frazer Institute, The University of Queensland, Brisbane

    Studying the topology of cancer tissue and its microenvironment to identify topological biomarkers (GNT2034141); identifying predictors of immunotherapy response in non-small cell lung cancer (MRF2031100); developing computational methods for high-plex, sub-cellular resolution spatial multiomics.

  • 2023 – 2024
    Senior Research Officer
    Computational Systems Oncology Group, SAiGENCI, The University of Adelaide

    Built analysis pipelines for spatial molecular technologies; identified spatial biomarkers in lung and head & neck cancers; studied genomic predictors of survivorship in paediatric brain cancers.

  • 2020 – 2023
    Research Officer
    Bioinformatics Division, Walter and Eliza Hall Institute (WEHI), Melbourne

    Spatial and single-cell transcriptomics of drug-induced phenotypic plasticity in cancer; multi-omics analysis of drug response; commercial research collaboration between the CRC for Cancer Therapeutics and Pfizer Inc.

  • 2016 – 2020
    PhD Candidate
    School of Mathematics and Statistics, The University of Melbourne

    Context-specific gene regulatory networks in cancer, single-sample molecular phenotype scoring, and several widely-used R packages. Supervised by Prof. Edmund Crampin and A/Prof. Melissa J. Davis.

Education

  • PhD, Mathematics and Statistics
    The University of Melbourne, 2020
  • MSc, Bioinformatics
    The University of Melbourne, 2015
  • BSc, Computer Science
    University of Southampton, 2013

Awards & recognition

  • Outstanding Early Career Researcher (ECR) Award
    Australian Bioinformatics and Computational Biology Society (ABACBS), 2025
  • NHMRC Emerging Leadership Fellowship
    National Health and Medical Research Council, 2025–2029
  • Best Written Lay Summary Award
    Victorian Cancer Bioinformatics Symposium, 2021
  • Melbourne Research Scholarship
    The University of Melbourne, 2016
  • MSc (Bioinformatics) merit-based bursary
    The University of Melbourne, 2014

Get in touch

Interested in spatial omics methods, collaborations, or graduate research supervision? I'd love to hear from you.

Email me