USC researchers have developed a new AI model to predict how proteins may bind to DNA. The model could be used to expedite the development of novel drugs and other treatments that are traditionally time-intensive processes.
The tool, published in Nature Methods, is called Deep Predictor of Binding Specificity (DeepPBS) and is a geometric deep learning model that can predict protein-DNA binding.
"Structures of protein–DNA complexes contain proteins that are usually bound to a single DNA sequence. For understanding gene regulation, it is important to have access to the binding specificity of a protein to any DNA sequence or region of the genome," said Remo Rohs, professor and founding chair in the Department of Quantitative and Computational Biology at the USC Dornsife College of Letters, Arts and Sciences. "DeepPBS is an AI tool that replaces the need for high-throughput sequencing or structural biology experiments to reveal protein–DNA binding specificity."
Being a geometric deep learning model allows DeepPBS to analyze data using geometric structures by capturing the chemical properties and geometric contexts of protein DNA binding to predict the binding specificity. Additionally, DeepPBS can predict binding specificity across multiple protein families.
"It is important for researchers to have a method available that works universally for all proteins and is not restricted to a well-studied protein family. This approach allows us also to design new proteins," Rohs said.
The researchers feel the applications of DeepPBS are countless. The tool could potentially be used to accelerate drug design, as well as lead to new discoveries in synthetic biology and RNA research.