Once a bustling biscuit factory, an old building in South London now hums with the rhythm of robotic arms, incubators, and DNA sequencing machines. The ovens and mixers have given way to a revolutionary process of engineering medical antibodies, all thanks to James Field and his company, LabGenius.
In our bodies, antibodies are the immune system's soldiers, fighting disease with their specially shaped protein strands. Since the 1980s, synthetic antibodies have been used to treat diseases like cancer and to prevent organ transplant rejection. However, the human process of designing these antibodies is like finding a needle in a haystack of amino acids.
Enter LabGenius. In 2012, Field, while pursuing his PhD in synthetic biology at Imperial College London, saw the falling costs of DNA sequencing, computation, and robotics as an opportunity. His company now uses these three elements to automate the antibody discovery process.
At their lab, a machine learning algorithm designs antibodies to target specific diseases. Robotic systems then build and grow these antibodies, run tests, and feed the data back into the algorithm, all with minimal human supervision.
The process starts with human scientists identifying a range of potential antibodies for a specific disease. LabGenius' machine learning model then explores this range, quickly and effectively. The model selects over 700 initial options from a search space of 100,000 potential antibodies. These are then automatically designed, built, and tested, with the aim of finding potentially fruitful areas to investigate in more depth.
The tests are almost entirely automated, with technicians overseeing the process. The experimental data is then fed back into the machine learning model, refining its understanding of the space. With each subsequent round of antibody designs, the model improves, balancing the exploitation of potentially fruitful designs with the exploration of new areas.
This innovative approach yields unexpected solutions, finds them more quickly, and could revolutionise the 'artisanal' process of drug discovery. As Field puts it, "You find molecules that you would never have found using conventional methods." The outcome? More effective antibody treatments with fewer side effects, leading to better patient outcomes.
So, next time you're enjoying a biscuit, spare a thought for the old biscuit factory in South London. It's no longer cooking up sweet treats, but instead, it's brewing a revolution in medical science, all thanks to the power of AI.
Made with TRUST_AI - see the Charter: https://www.modelprop.co.uk/trust-ai
Comments