Matthew A. Clarke

I am a Research Fellow at PIBBSS (Principles of Intelligent Behaviour in Biological and Social Systems). My current research focuses on the mechanistic interpretability of machine learning, specifically using sparse autoencoders. This work combines my interests in computational modeling and complex systems, now applied to understanding the inner workings of AI.

I was previously a postdoc working at the Cancer Institute at University College London. My main research interests combine biology and computer science, using computational modelling to predict cancer evolution and plan treatment programmes to avert or overcome resistance.

I completed my PhD at the the University of Cambridge, looking at how computational network models could be used to find more effective combination treatments for breast cancer. As a postdoc at the Fisher Lab in the UCL Cancer Institute I built upon this work in order to predict resistance mechanisms to radiotherapy and to find the most effective patient-specific treatments to overcome them.

I am keen to share my knowledge and expertise with others. As a mentor to postdocs, PhD students, Masters students, and undergraduates, I take pride in helping to guide and inspire the next generation of scientists. If you have any questions about my work, please get in touch with the contact form below.

Experience

 
 
 
 
 
Principles of Intelligent Behavior in Biological and Social Systems (PIBBSS)
Research Fellow
June 2024 – Present London, UK
My current research focuses on the mechanistic interpretability of machine learning, specifically using sparse autoencoders. This work combines my interests in computational modeling and complex systems, now applied to understanding the inner workings of AI. I am being mentored by Joseph Bloom.
 
 
 
 
 
Cancer Institute, University College London
Associate Staff Visitor
June 2024 – Present London, UK

I am continuing my work on triple-negative breast cancer, non-small cell lung cancer and the automation of network generation using message-passing graph neural networks in the Jasmin Fisher lab.

Responsibilities included:

  • Building network models by collating data on biochemical interactions in collaboration with experimental partners.
  • Development of new methods for the generation and analysis of executable network models.
  • Developing and maintaining research software developed and used by the lab.
 
 
 
 
 
Cancer Institute, University College London
Postdoctoral Research Fellow
April 2020 – June 2024 London, UK

As part of the Jasmin Fisher lab, I continued my work on DNA damage repair, applying this to breast cancer and lung cancer. I also developed new methods for studying cancer evolution. During the pandemic, my colleagues and I demonstrated how our work on cancer could be applied to infectious disease, to predict drug repurposing for rapid response to the COVID-19 epidemic.

Responsibilities included:

  • Building network models by collating data on biochemical interactions in collaboration with experimental partners.
  • Development of new methods for the generation and analysis of executable network models.
  • Supervision of undergraduate and Master’s students.
  • Interviewing hiring candidates.
  • Mentoring new members of the lab.
  • Presenting at international conferences.
  • Developing and maintaining research software developed and used by the lab.
  • Lab Management.
 
 
 
 
 
Promatix Biosciences Ltd
Consultant
October 2021 – September 2023 London, UK

I worked to support Promatix Ltd to accelerate the hunt for oncology therapeutics as part of their collaboration with the Jasmin Fisher lab.

Responsibilities included:

  • Interviewing hiring candidates.
  • Advising on research.
 
 
 
 
 
Department of Biochemistry, University of Cambridge
Postdoctoral Research Associate
October 2018 – April 2020 Cambridge, UK

With funding from CRUK RadNet, I worked in the Jasmin Fisher lab on modelling the DNA damage response pathway to find radiosensitising drug treatments. I also continued my work on breast cancer, focussing on the triple-negative sub-type (TNBC) as part of a collaboration with the Mark Foundation for Cancer Research and PARTNER clinical trial. In collaboration with Jasmin Fisher I published a review on the opportunities and challenges for executable modelling.

Responsibilities included:

  • Building network models by collating data on biochemical interactions in collaboration with experimental partners.
  • Supervision of undergraduate and Master’s students.
  • Presenting at international conferences.
  • Mentoring new members of the lab.
 
 
 
 
 
Department of Biochemistry, University of Cambridge
Visiting Scholar
January 2018 – October 2018 Cambridge, UK

Thanks to generous funding from the Glover Fund, I was able to build upon my work on the evolution of cancer, expanding it to understand how the behaviour of blood cancers is determined by the order in which they acquire mutations, even when their final mutational profile is identical as part of a collaboration of the Jasmin Fisher lab with the Hall and Kent labs.

Responsibilities included:

  • Building network models by collating data on biochemical interactions in collaboration with experimental partners.
  • Supervision of undergraduate and Master’s students.
  • Presenting at international conferences.
 
 
 
 
 
Department of Biochemistry, University of Cambridge
PhD in Computational Biology
October 2013 – January 2018 Cambridge, UK

As part of the Wellcome Trust Mathematical Genomics and Medicine PhD programme, studied as part of the University of Cambridge Computational Biology MPhil, while also undertaking rotations in different labs. For my PhD I was supervised by Jasmin Fisher in the University of Cambridge Department of Biochemistry and Microsoft Research Cambridge, and my advisor was Trevor Littlewood. During my PhD, I developed a network model of breast cancer, as well as novel methods to study combination treatment and the evolution of cancers. Using these, in collaboration with the Gerard Evan lab, we were able to identify and validate a novel combination treatment for breast cancer, exploiting the heterogeneity observed in myc driven breast cancers. My PhD examiners were Bertie Göttgens and Francesca Buffa.

Responsibilities included:

  • Building network models by collating data on biochemical interactions in collaboration with experimental partners.
  • Supervision of undergraduate and Master’s students.
  • Presenting at international conferences.
 
 
 
 
 
Department of Physics, University of Cambridge
Masters in Experimental and Theoretical Physics
October 2012 – October 2013 Cambridge, UK
For my Master’s thesis, I studyed the how fibroblasts sense and respond to mechanical changes in their environment, in particular in wound healing. In the Terentjev lab worked on stability analysis of non-linear differential equations modelling this system, particularly focussing on Lyapunov Methods.
 
 
 
 
 
University of Cambridge
BA (MA Cantab) in Natural Sciences
October 2009 – October 2012 Cambridge, UK
Reading Natural Sciences gave me the opportunity to build skills in diverse areas of science and mathematics, ranging from particle physics to experimental biology. This gave me the curiosity about how the programming underlying cell behaviour functions, and the mathematical and computational skills to pursue those questions.

Projects

*
SAE Latent Co-occurrence
Sparse AutoEncoder (SAE) latents show promise as a method for extracting interpretable features from large language models (LM), but their overall utility as the foundation of mechanistic understanding of LM remains unclear. Ideal features would be linear and independent, but we show that there exist SAE latents in GPT2-Small display non-independent behaviour, especially in small SAEs. We are investigating: 1. What fraction of SAE latents might best be understood in groups rather than individually? 2. Do low-dimensional subspaces mapped by co-occurring groups of features ever provide a better unit of analysis (or possibly intervention) than individual SAE latents?
Predicting Cancer Evolution
Cancers evolve from normal cells, and continue to evolve to evade treatment. Understanding evolution from tumourigenesis will allow for earlier detection of cancer, while predicting evolution of resistance mechanisms is essential for finding treatments that are effective in the long term.
Combination therapy
Computatational modelling allows screening of thousands of combinations of drugs to find the most effective treatments. This allows for better protection against the emergence of resistance, as well as rapid drug repurposing e.g. to combat emerging diseases such as COVID-19.
DNA Damage Repair and Radiotherapy
Cancers often emerge, in part, due to deficencies in DNA damage repair, making them vulnerable to DNA damaging treatments such as radiotherapy. While radiotherapy efficacy has been increasing, this is mainly due to better targeting of tumour anatomy, but targeted therapy offers the opportunity to radiosensitise tumour cells by targeting vulnerabilities in the tumour biology. We use computational modelling to find such vulnerabilities, and explore potential resistance mechanisms.

Recent Publications

Quickly discover relevant content by filtering publications.
Executable Network of SARS-CoV-2-Host Interaction Predicts Drug Combination Treatments
The COVID-19 pandemic has pushed healthcare systems globally to a breaking point. The urgent need for effective and affordable COVID-19 …

Recent & Upcoming Talks

Predicted combination treatments for NSCLC using in silico CRISPR screening
Lung cancer is the leading cause of cancer mortality world-wide, with over half of lung cancer patients relying on radiotherapy (RT). …
Predicted combination treatments for NSCLC using in silico CRISPR screens
Lung cancer is the leading cause of cancer mortality world-wide, with over half of lung cancer patients relying on radiotherapy (RT). …

Contact

  • 25 Holywell Row, London, EC2A 4XE