Research and Training to create assets in the form of new companies to develop a new model to generate innovative therapies

Lead Research Organisation: Institute of Cancer Research
Department Name: Division of Molecular Pathology

Abstract

Cancer Research UK has partnered with Deep Science Ventures to create an opportunity for early career stage researchers to build ventures that target the complexity of biology. This partnership brings together the world's largest independent funder of cancer research with the unique venture design process of Deep Science Ventures to create a programme that is exhaustive in the search for the best approach from both technical and commercial perspectives, and positioned for new venture growth from day one. Their aim is to identify and bring together entrepreneurial scientists, academic advisors and investors to design and build new companies in oncology.

In this project, the research goal is to develop therapies by leveraging the wealth of data of this new era to identify ways to kill cancer cells so that can escape from treatments.

My role will be to start as a Founder Analyst within DeepScienceVentures under their unique framework for venture creation. It's a part-time role working mainly involving research. In order to make this research successful, I will apply my skills sets of machine learning/artificial intelligence, understanding heterogeneity, developing patient-based biomarkers and hands-on experience in preclinical trials. This also provides opportunity to get trained in market analysis, and customer development together with the DSV team before incorporating our own start-up.

Technical Summary

Cancer is an evolution machine. Decades before diagnosis, a multitude of different clones, carrying differential sets of mutations are setting the scene that makes almost any targeted therapy obsolete. The new era of personalised medicine and of immense knowledge of the genetics, epigenetics, gene expression, metabolomics, and clonal evolution of tumours, converges with the completion of flagship synthetic lethality studies, interrogating pan-cancer functional genomics. However, target discovery still focuses on finding targeted therapies against well-established cancer drivers from which cancer has already evolved out of, even years before diagnosis. This results in limited response rates in the clinic and quick evolution of resistance for targets against such drivers for example BRCA targeted PARP inhibitors.
Our Approach: We are stopping the chase for moving targets by building a highly advanced drug discovery machine, that leverages the wealth of data of this new era to identify vulnerable mechanisms that cancer cannot evolve out of and the specific context in which they are effective.

To achieve that, we identify cancer systems biology and specific mechanisms of irreversible cancer dependencies through advanced computational methods.

Our current approach sets the bar very high, to produce complete and sustained therapeutic responses. We have invented methods that have already produced potential candidates with strong independent validation.My role will be to start as a Founder Analyst within DeepScienceVentures under their unique framework for venture creation. It's a part-time role.

As a secondee, my goal is to join the host institution to create a company that will bring novel curative therapies to cancer patients.

In order to make this successful, I will apply my skills sets of machine learning/artificial intelligence, understanding heterogeneity, developing patient based biomarkers and hands-on experience in preclinical trials.

Publications

10 25 50
 
Title Machine learning approach to identify biomakers for drug targets 
Description I developed a machine learning approach to identify biomarkers based on a drug target and another genomic aberration. 
Type Of Material Computer model/algorithm 
Year Produced 2021 
Provided To Others? No  
Impact This is expected to potentially personalise drugs, if our target gets developed into a drug. 
 
Title Mixture model to identify drug targets 
Description I developed a mixture linear model to identify drug targets from cell lines. 
Type Of Material Computer model/algorithm 
Year Produced 2021 
Provided To Others? No  
Impact This is leading to the identification of cancer-specific drug targets 
 
Description CRUK Therapeutic Discovery Lab 
Organisation Cancer Research UK
Country United Kingdom 
Sector Charity/Non Profit 
PI Contribution As a Secondee to Enedra Therapeutics, a part of the Deep Science Venture, a part of my role is develop machine-learning models for drug target discovery.
Collaborator Contribution Through CRUK Therapeutic Drug Discovery Lab, we collaborate with them to validate our drug targets.
Impact Ongoing work.
Start Year 2021
 
Description KQ Labs - Accelerator Programme for early-stage startups 
Organisation Francis Crick Institute
Country United Kingdom 
Sector Academic/University 
PI Contribution As a Secondee to Enedra Therapeutics, a Spin-off from Deep Science Ventures, I participate to learn to validate ideas and business models.
Collaborator Contribution KQ Labs is a five-month accelerator programme for exciting early-stage start-ups using data as a core part of their business model to improve human health, run by the Francis Crick Institute and funded by LifeArc.
Impact NA
Start Year 2021