Healthy growth and development depend on tens of thousands of genes being activated at the right time and place. Specific regions of DNA help coordinate this process, guiding the production of enzymes, hormones, proteins, and other molecules that cells need to function properly. When gene activation fails, cells can malfunction and contribute to diseases, including cancer. To better understand the DNA sequences that control this process, researchers at the University of California, San Diego’s laboratory, Professor James T. Kadonaga, focused on an important element of DNA known as the “initiator.” The primer marks the place where the information encoded in a gene begins to be converted or expressed into a functional product. AI decodes primer sequence In the new study, led by graduate student researcher Torrey Rhyne-Carrigg, the team used high-throughput DNA sequencing to measure gene expression activity in approximately 500,000 different versions of the primer. The researchers then used those results to train a machine learning system, a form of artificial intelligence, to identify the characteristic DNA pattern associated with the starter. Once the model had decoded that signature, the team searched human genes for the sequence and found that about 60% contain the primer. “These AI models were found to provide, for the first time, robust predictions of the presence or absence of the primer in human genes and were therefore able to decode the base sequence pattern of the primer DNA,” said Kadonaga, a professor in the School of Biological Sciences in the Department of Molecular Biology at UC San Diego. Predicting the effects of DNA mutations The findings could help researchers anticipate how mutations that affect the initiator can alter gene activity and contribute to a variety of disorders. The study data and AI models can also support the design of synthetic promoters, sequences that can turn genes on or off, with functions designed for specific purposes. More broadly, the research shows how laboratory experiments and artificial intelligence can be combined to discover information encoded in human DNA. “On a more global level, this work is a step forward in the combined use of laboratory experiments and artificial intelligence to decipher the information contained in the sequence of DNA bases in humans,” said Kadonaga. “Ultimately, within the six billion bases of DNA in each of our cells, there is a gene expression code that specifies when, where, and to what extent each of our genes should be turned on or off. If we had an AI model for the entire gene expression code, we could predict the activity of each of the different gene variants in different people. The new AI model for the starter is a small but important part of this gene expression code, and I am optimistic that we will expand our AI models of the being. human gene expression code in the not too distant future”.
