Andrew Wood

Genetic technologies to control proteins, cells and tissues.

Andrew Wood is a Reader and Group Leader at the MRC Human Genetics Unit in the Institute of Genetics and Cancer. His research uses genetic engineering technologies to advance our understanding of human disease and therapy. In particular, he aims to understand how drugs and biologics interact with cells and tissues. Current projects focus on genome editing nucleases and targeted protein degraders.

Andrew did his PhD at Kings College London, where he identified new mechanisms of epigenetic gene regulation working with Rebecca Oakey. He then moved to UC Berkeley as a Sir Henry Welcome Fellow, where he joined Barbara Meyer and scientists at Sangamo Biosciences to apply ZFN and TALEN genome editors across animal species. 

portrait photo of Andrew Wood
Andrew Wood

Andrew moved to Edinburgh in 2011, first joining Wendy Bickmore’s group before starting his own laboratory supported by a Sir Henry Dale Fellowship in 2014. 

Gillian Taylor, Abram Giller, Emma Ramsey, and Evelina Gudauskaite


Targeted protein degradation

Degrader molecules such as PROTACs and molecular glues rapidly destroy their protein targets via the ubiquitin proteasome system. These tools are being used by an increasing number of research groups to address fundamental questions in cell biology while, at the same time, a wave of degrader therapeutics is progressing through clinical trials. In collaboration with the MRC National Mouse Genetics Network (https://nmgn.mrc.ukri.org/), we have built synthetic biosensor proteins to better understand the pathways through which degraders access and destroy target proteins in mammalian tissues. Several collaborative projects use genome editing and protein tagging to validate degrader targets in specific disease areas (e.g. Macdonald et al, 2022). We develop strategies for tagged protein design that minimise disruption of normal protein function (e.g. Taylor et al, 2025). 

Genome editing

We developed a sensitive system to understand how different chromatin states affect the frequency and mutagenic outcome of Cas9-mediated mutagenesis in mammalian stem cells (Kallimasioti et al, 2018). By targeting imprinted genes, we found that transcriptionally silent alleles accumulated mutations at a much slower rate, but showed no difference in the rate of precise repair from donor templates. 

We and others have developed methods that use CRISPR to make hundreds of different point mutations at a single genomic locus in a population of cells and simultaneously measure their function. In collaboration with Peter Hohenstein and Derya Ozdemir, we used this saturation genome editing approach to understand the consequences of all possible missense mutations spanning one of the most frequently mutated regions of the human cancer genome, within the oncogene and regulator of wnt signalling, β-catenin (Krishna et al, 2025). These mutations prevent β-catenin from being broken down, allowing it to build up and activate growth-promoting genes. Using a fluorescent signalling test, we created a detailed “map” showing how strongly each mutation activates the Wnt pathway. This revealed that common cancer mutations differ widely in strength: some cause only weak activation, while others lead to very strong signalling.

By comparing these results with tumour data from thousands of cancer patients, we found that different tissues are enriched for mutations that cause different levels of β-catenin activity. In liver cancer (hepatocellular carcinoma), two main groups of tumours emerged: one with weakly activating CTNNB1 mutations and another with strongly activating ones. Importantly, the weaker mutations were linked to greater immune cell infiltration, suggesting that the specific level of β-catenin activation may influence how tumors interact with the immune system and respond to immunotherapy.


Figure showing targeted degradation using a degron tag

Figure legend

  1. Schematic diagram illustrating the steps of degron tagging. A genetic cassette encoding a ligand binding site is fused with an endogenous gene of interest to produce a tagged protein. When cells expressing this protein are exposed to a degrader ligand recognising the tag, an E3 ubiquitin ligase complex is recruited to ubiquitinate the tagged protein, leading to rapid destruction via the proteasome.
  2. Heatmap shows the level of wnt signalling activation resulting from all possible single amino acid substitutions across codon positions spanning the β-catenin degron. Red indicates high signalling, blue = low. The lower histogram shows the frequency of missense substitutions at each codon position among tumours recorded in the COSMIC human cancer database. 

Krishna A, Meynert A, Kelder M, Ewing A, Sheraz S, Ferrer-Vaquer A, Grimes G, Becher H, Silk R, Semple C, Kendall T, Hadjantonakis A, Bird T, Marsh JA, Hohenstein P*, Wood AJ*, Ozdemir D* (2025) Mutational scanning reveals oncogenic CTNNB1 mutations have diverse effects on signalling. Accepted, Nature Genetics. Preprint: 10.1101/2023.11.09.566307v1

Taylor G, Macdonald L, Szulc N, Gudauskaite E, Hernandez Moran B, Brisbane JM, Donald M, Taylor E, Zheng D, Gu B, Mill P, Yeyati PL, Pokrzywa W, Ribeiro de Almeida C, Wood AJ. (2025) Tissue-specific consequences of tag fusions on protein expression in transgenic mice. PLoS Genetics https://doi.org/10.1371/journal.pgen.1011830

Macdonald L, Taylor G, Brisbane J, Christodoulou E, Scott L, Von Kriegsheim A, Rossant J, Gu B*, Wood AJ* (2022) Rapid and specific degradation of endogenous proteins in mouse models using auxin-inducible degrons. eLife https://doi.org/10.7554/eLife.77987

Kallimasioti-Pazi I, Chathoth K, Taylor G, Meynert A, Ballinger T, Kelder M, Sanli I, Lalevee S, Feil R, Wood AJ. (2018) Heterochromatin delays CRISPR-Cas9 mutagenesis but does not influence the outcome of mutagenic DNA repair. PLoS Biology 16:e2005595

#Wood AJ, #Lo TW, #Zeitler B, Pickle CS, Ralston EJ, Lee AH, Amora R, Miller JC, Leung E, Meng X, Zhang L, Rebar EJ, Gregory PD, Urnov FD, Meyer BJ. (2011) Targeted genome editing across species using ZFNs and TALENs. Science. 333:30
* co-senior author, #co-first author


This article was published on