On July 27, 2023, the mid-term evaluation meeting for the National Key Research and Development Program of China (NMR) project "Research and Application of Artificial Intelligence Algorithms for Massive Engineering Trial and Error Optimization of Synthetic Biological Systems," chaired by Professor Li Jianqiang, Executive Director of the National Engineering Laboratory, was successfully held in Room 1007 of Zhizhen Building.
Attendees included Professor Li Jianqiang, Executive Director of the National Engineering Laboratory; Zhang Jiefeng, Deputy Director of the Science and Technology Department of Shenzhen University; Zhi Li, a manager from the National Natural Science Foundation of China (NSFC); Huang Yanting from the Finance Department; and Professor Deng Li from the College of Life Sciences and Oceanography, along with all key members of the project team.
Professor Li Jianqiang delivered an online PPT presentation, detailing the project team's research over the past two and a half years from six aspects: project background and scientific questions, research content and expected goals, research team and personnel input, project progress and research results, fund utilization, existing problems, and future work plans. Experts from the Ministry of Science and Technology affirmed the project team's research work and asked questions regarding the project's completion and future development, successfully completing the mid-term evaluation.
The project "Research and Application of Artificial Intelligence Algorithms for Massive Engineering Trial and Error Optimization of Synthetic Biological Systems" addresses the problem that traditional mathematical models are insufficient for effectively designing complex synthetic biological systems. It studies the deep integration of massive engineering trial and error optimization of synthetic biological systems with artificial intelligence technology, constructs a semantically defined and dynamically updated symbiotic knowledge base for synthetic biological systems, establishes a massive engineering trial and error optimization platform based on reinforcement learning and other models with continuous learning capabilities, and combines high-throughput automated construction and multi-spectral characterization techniques to improve the predictive and characterization capabilities of deep learning-based synthetic biological systems. The project has been validated in specific scenarios such as transcriptional regulation, enzyme catalysis, and immune signal transduction, completing a closed-loop design of "design-construction-testing-learning" based on artificial intelligence. The research focuses on four objectives: (1) constructing a semantically defined and dynamically updated symbiotic knowledge base for synthetic biological systems; (2) establishing a massive engineering trial and error optimization platform based on artificial intelligence; (3) developing high-throughput automated genotype and phenotype analysis experimental techniques; and (4) improving the accuracy and interpretability of deep learning-based synthetic biological system prediction models.
