New research published in the journal Nature shows that artificial intelligence (AI), when combined with advanced experimental science, can help uncover previously unknown proteins in the human body and determine their roles. Researchers at Sylvester Comprehensive Cancer Center, part of the University of Miami Miller School of Medicine, used AI to analyze the three-dimensional shapes of over 214 million predicted proteins. Instead of looking at genetic sequences, they focused on protein structures, identifying new members of the G protein-coupled receptor (GPCR) family. These receptors are crucial for cells to detect and respond to signals from their environment, such as hormones or neurotransmitters. One of the newly discovered proteins, called TM184C, caught the researchers’ attention. While it resembles a GPCR in structure, it behaves differently. TM184C is typically found inside cells, embedded in the membranes of small sacs called intracellular vesicles. These vesicles move along structures known as microtubules and gather at thin extensions that link neighboring cells. These extensions, known as tunneling nanotubes, act as bridges, allowing cells to exchange substances like nutrients, organelles, and even mitochondria—the energy-producing parts of cells. When researchers disrupted TM184C, the cells formed fewer connections and showed changes in their shape and the organization of their vesicles. This suggests that TM184C plays a role in building and managing these intercellular bridges. The discovery raises intriguing questions about whether these exchanges benefit all cells equally or if some cells gain at the expense of others. In addition to its role in cell communication, TM184C also appears to regulate autophagy, the process by which cells break down and recycle old or damaged components. When TM184C levels were reduced, markers of autophagy increased, indicating that the protein helps control this process. The researchers also studied a yeast protein called Hfl1, which is similar to TM184C. When Hfl1 was removed from yeast cells, they experienced problems, but adding the human TM184C protein restored normal function. This suggests that the role of this protein has remained largely unchanged in yeast and humans, despite over a billion years of evolution. The study underscores the value of combining AI with traditional experimental methods. While AI can significantly speed up scientific discovery, researchers caution that it should not be used blindly. Instead, it should be guided by experts who can validate and interpret the findings. This approach is not only accelerating research but also revealing new aspects of cellular behavior, including how cells communicate, respond to stress, and contribute to disease.