.png){kind=logo}
Google’s C2S-Scale 27B foundation model, developed in partnership with Yale University and based on the Gemma framework, has predicted a new hypothesis of cancer cell behavior. Interestingly, the hypothesis was also verified experimentally in living human cells.
Google CEO Sundar Pichai referred to the AI-assisted laboratory science milestone as a monumental breakthrough. He pointed out that this landmark can influence both preclinical and clinical research, and potentially open up new therapeutic avenues in the future.
The artificial intelligence model was educated on more than one billion single-cell profiles and translates molecular information into ‘cell sentences,’ enabling it to read cellular communication. This proficiency helps the system to acquire the ‘language of cells’, reason about intricate biological interactions, and develop hypotheses in place of summarizing what is already known.
Biotech entrepreneur Bryan Johnson described that the model literally ‘listened’ for signals cells were communicating before it made a prediction validated by subsequent experiments.
The model predicts a considerable enhancement of about 50% in the antigen presentation of cancer cells when silmitasertib (CX-4945) is used together with low interferon. This is the very process through which immune cells can identify tumor cells, potentially turning the previously ‘invisible’ cold tumors into those visible to the immune system.
The prediction was validated through experiments on human cell models, which hold promise for novel approaches to immunotherapy for cancers resistant to standard therapies.
The milestone that has been accomplished is proof of AI’s capacity to generate novel scientific hypotheses and to couple computer-based predictions with experimental validation, which has changed the course of biomedical research.
The development still points to the AI’s potential to speed up the process of innovation and to guide research in ways that are not seen before, although extensive preclinical and clinical testing are still required before patient treatments can come out.
Also Read: Nano Banana AI Model Expands to Google Search, NotebookLM, and Photos
Google has released the C2S-Scale model to the public as part of its strategy to democratize AI-powered biomedical research. The model can now be accessed on platforms like Hugging Face and GitHub, which is a great opportunity for scientists all over the world to interact with the model and validate its predictions. This, in turn, will lead to the expansion of its capabilities, thus facilitating joint efforts in the fight against cancer.
