Robots, computers, and algorithms have learned to process huge amounts of data and create antibody molecules that were previously unimaginable by humans. This breakthrough opens up new prospects in the search for potential methods of treating known diseases (source).
In the southern part of London, on the site of the old cooking factory, Labgenius has established a modern laboratory. Instead of traditional machinery and industrial furnaces, the lab now houses robotic equipment, incubators, and DNA sequencers.
In 2012, James Field, the founder and CEO of Labgenius, saw the potential for cost reduction in DNA sequencing, computing power, and robotics. This inspired him to automate the process of creating new medical antibodies using these technologies.
In nature, antibodies are the immune system’s response to infection. These proteins attach to foreign substances and remove them from the body. Since the 1980s, pharmaceutical companies have been developing synthetic antibodies to treat cancer and prevent organ rejection. However, the development of these antibodies has been a slow process for humans.
Labgenius has created a machine learning algorithm that significantly accelerates the search for necessary antibodies. According to James Field, all a person needs to do is provide an example of a healthy and sick cell, and the system will explore millions of potential combinations of amino acids to find the most suitable one.
The process is almost completely automated – robots grow antibodies based on their genetic sequence, test them on samples of sick cells, and then transmit the results back to the algorithm. Humans are only responsible for moving samples between stages. This machine training allows for the discovery of unexpected solutions that humans may not have considered at all.
Field believes that this approach can produce completely unique and even conventional antibodies that will provide the best therapeutic effect for patients.
This automated approach reduces the development time for antibodies from several years to just 6 weeks, paving the way for the creation of more effective and safe drugs. By automating routine tasks, researchers can quickly test a much larger number of options and find optimal solutions that may be overlooked by human intuition. This method is expected to play a crucial role in improving medicine in the future.