Scientists developed a new artificial intelligence model designed to understand how genes interact and function together inside human cells. The system, called a Gene Set Foundation Model (GSFM), was inspired by large language models such as ChatGPT. Just as language models learn how words change meaning depending on context, the GSFM learns how genes behave differently across various biological conditions and cellular environments. Researchers believe this technology could significantly improve the understanding of complex diseases and cellular behavior.
The article explains that genes rarely work independently. Instead, they operate in interconnected networks that change depending on the cell type, disease state, or biological process involved. To train the AI system, researchers compiled millions of gene sets from scientific studies and gene expression datasets collected across thousands of experiments. The AI learned patterns by predicting missing genes within these sets, gradually building a sophisticated understanding of how genes cooperate and influence one another in different contexts.
One of the most important achievements of the model is its ability to identify relationships between genes before those relationships are experimentally confirmed in laboratories. During testing, the AI successfully predicted gene-gene and gene-function interactions that later appeared in newly published scientific studies. Researchers say this capability could help scientists identify poorly understood genes, discover new disease biomarkers, and uncover potential drug targets more efficiently than traditional experimental methods alone.
The study reflects a broader trend in biology where AI is increasingly being used to analyze massive biological datasets that humans cannot easily interpret on their own. By creating a reusable knowledge framework for understanding gene interactions, the GSFM may improve diagnostics, precision medicine, and biomedical research across multiple diseases. Scientists believe tools like this could eventually help researchers better understand how cells function, how diseases develop, and how treatments can be designed more effectively using AI-driven biological insights.
