Artificial Intelligence Accelerates Medical Development: “A Revolution on the Scale of the Discovery of Electricity”

Immunotherapy, a method that stimulates the immune system to fight diseases, is becoming increasingly important in modern medicine. Expert forecasts indicate that the market for antibodies used in therapy may grow threefold over the next decade, according to an analysis by Fact.MR. In Poland, the start-up Genotic is attracting attention with its achievements, using artificial intelligence (AI) to design antibodies, which shortens the process from several months to just 21 days.

Genotic founder Grzegorz Warzecha emphasizes that artificial intelligence opens up new possibilities in biotechnology and pharmacy, enabling automation of many processes, including the search for new drugs, especially more complex ones, such as antibodies. “Antibodies are among the most effective drugs, but their design is highly demanding. Thanks to deep learning networks, it is becoming possible to develop even personalized drugs for specific patients,” says Warzecha. For the medical industry, this is a revolutionary change, comparable to the discovery of electricity.

Startup Genotic, with the help of AI, develops antibodies for the diagnosis and treatment of infectious, autoimmune and oncological diseases. This process involves predicting the structure of proteins based on genetic sequences, which significantly speeds up the creation of antibodies directed at specific antigens.

“Our goal is to create a broad catalog of antibodies targeting different targets, such as HER2 in breast cancer. We are also working on automating laboratories using deep learning networks, which can significantly speed up research and achieve better results in less time,” Warzecha adds.

One of the key examples of AI in medicine is AlphaFold 3, a model created by Google DeepMind and Isomorphic Labs. This technology enables modeling of large biological molecules such as proteins, DNA, and RNA, as well as their chemical modifications, which is of great importance in developing new therapies. Genotic has incorporated AlphaFold into its platform, using it to generate the structure of antibodies.

According to Warzecha, the biggest advantage of deep learning networks is the ability to conduct research processes entirely digitally, so-called “in silico”, which simplifies the search for drug candidates. In the traditional approach, antibody design is based on long-term experiments conducted on animals. Only after obtaining an immune response is it possible to sequence and produce antibodies. Genotic, on the other hand, is able to design an antibody within 48 hours, using 200-300 graphics cards, which significantly reduces the time needed to verify candidates.

The effectiveness and specificity of the selected antibodies are then assessed, and from among thousands of potential candidates, several with the best properties are selected for further testing.

Monoclonal antibodies are already a key element of therapy, especially in the treatment of cancers such as multiple myeloma and DLBC lymphoma. Thanks to their bispecific properties, they are able not only to recognize abnormal cells, but also to teach the immune system how to fight them. The market for these antibodies is growing rapidly – in 2022 it reached a value of over $197 billion, and by 2032 it may exceed $608 billion.

Warzecha believes that the future of patient treatment will be related to creating personalized medicines based on the analysis of patients’ RNA and DNA. Pharmaceutical companies will be able to develop and deliver such therapies in a short time, which is a new, rapidly developing trend around the world.