As announced by Sundar Pichai, Google’s AI, in collaboration with Yale University, developed a novel hypothesis for cancer treatment.

Following the recent developments coming to reality with the advent of AI, the announcement by Alphabet’s CEO Sundar Pichai comes at the right time to underscore the possibilities that the very presence of  AI  confirms. Referred to as “an exciting milestone for AI in science” by Pichai, this Google’s C2S-Scale 27B foundation model, developed in collaboration with Yale University and based on the Gemma framework, generated a novel hypothesis about cancer cellular behaviour, which scientists experimentally validated in living cells.

The Hypothesis

The hypothesis suggested the drug Silmitasertib could significantly boost antigen presentation in cancer cells by almost 50% under conditions of low interferon. Antigen presentation is the process by which tumor cells display fragments of abnormal proteins on their surface, signaling the immune system to recognise and attack them. 

This prediction by AI was confirmed by the experiment conducted in human cell models. If these results are validated in further preclinical and clinical studies, they could open the door to new immunotherapy approaches, offering fresh hope for patients with cancers that are notoriously difficult to treat.

Read more | Hyderabad Airport Unveils India’s First AI Food Delivery Robot

How is it possible?

At the centre of this achievement lies a complex procedure of training the AI system on over one billion single-cell profiles and transforming raw molecular data into what researchers call “cell sentences”. Through this procedure, AI was able to interpret cellular communication, essentially learning the language of cells. Explained by biotech entrepreneur Bryan Johnson, the AI model “listened” to what cells were signalling and produced a hypothesis. 

This marks a turning point for the involvement of AI in medical research, as AI is now capable of creating an entire new scientific prediction rather than just analysing the pre-existing data set.