The article reports on a study conducted by researchers at the University of California San Diego, led by Professor James T. Kadonaga and graduate student Torrey Rhyne-Carrigg, which focused on decoding a DNA element known as the “initiator.” The initiator marks the location where gene information begins to be expressed into functional products. Using high-throughput DNA sequencing, the team analyzed approximately 500,000 versions of the initiator to train a machine learning system to identify its characteristic DNA pattern. The AI model successfully decoded the initiator’s signature, revealing that roughly 60% of human genes contain this element.
The researchers suggest that the AI models’ ability to predict the presence or absence of the initiator in human genes could help anticipate how mutations affecting the initiator might alter gene activity and contribute to various disorders. Additionally, the study’s data and AI models may support the design of synthetic promoters for specific gene regulation purposes. The findings demonstrate the potential of combining laboratory experiments with artificial intelligence to uncover information encoded in human DNA, representing a step toward deciphering the broader gene expression code that specifies when, where, and to what extent genes are activated or deactivated. Professor Kadonaga expressed optimism that the new AI model for the initiator is a small but significant part of this larger gene expression code, with plans to expand AI models to cover more aspects of human gene expression in the future. [Source: https://www.sciencedaily.com/releases/2026/08/260823014943.htm]