With generative AI, MIT chemists quickly calculate 3D genomic structures
Peer-Reviewed Publication
Updates every hour. Last Updated: 1-Jul-2025 00:10 ET (1-Jul-2025 04:10 GMT/UTC)
MIT chemists found a new way to determine 3D genome structures, using generative AI, that can predict thousands of genome structures in minutes, making it much speedier than existing methods for analyzing the structures.
the results of a new study challenge the prevailing belief that some breeds are inherently superior at specific tasks, based on their skull morphology. The study, led by Nicholas Hebdon and Lindsay Waldrop and published in Science Advances on January 29, used advanced 3D reconstruction techniques to analyze 117 skulls from 40 domestic dog breeds and 18 wild canid species. The researchers found substantial overlap in skull shapes across breeds and functional categories, but no clear evidence that breeds selected for bite work or scent work have developed distinct morphological traits that enhance these abilities. This suggests that humans have been breeding dogs primarily for preferred visible traits, and that other factors like individual personality affect dogs’ performance of tasks.