China’s Tianhe-2 supercomputer, ranked among the top-10 fastest computers globally, has demonstrated potential in enhancing drug discovery through a Sino-US research collaboration. Scientists from Sun Yat-sen University, Beijing-based AI startup Galixir, and researchers from the Georgia Institute of Technology and the Massachusetts Institute of Technology developed a deep-learning-based toolkit called BioNavi-NP. This innovative tool predicts biosynthetic pathways for natural products (NPs) and NP-like compounds, addressing challenges in the synthesis and development of NPs, which constitute over 60% of FDA-approved small-molecule drugs.
Research Collaboration and Publication
The collaborative effort resulted in the publication of “Deep learning driven biosynthetic pathways navigation for natural products with BioNavi-NP” in Nature Communications. The study highlights the application of advanced computational methods to streamline the identification and development of natural products as drug candidates.
BioNavi-NP Toolkit
BioNavi-NP is a single-step bio-retrosynthesis prediction model trained using end-to-end transformer neural networks. It generates candidate precursors for target NPs by leveraging both general organic and biosynthetic reactions. The toolkit successfully identified biosynthetic pathways for 90.2% of 368 test compounds, demonstrating its high accuracy and utility in the field.
Web Server and Accessibility
The researchers have made the BioNavi-NP web server publicly available. It allows users to predict biosynthetic pathways and assess the biological feasibility of these pathways based on estimated species and enzyme preferences, providing a valuable resource for the scientific community engaged in natural product research.-Fineline Info & Tech
