生命序列系统的层级嵌套结构和算法信息论

：刘宇（北京师范大学珠海校区）
：2024-06-05 15:00
：海韵园数理大楼686会议室

报告人：刘宇（北京师范大学珠海校区）

时  间：2024年6月5日15:00

地  点：海韵园数理大楼686会议室

内容摘要:

In life systems, the evolution of proteins, genes, secondary metabolites, and others in their natural state never starts from scratch but always involves "tinkering." This means new entities are produced by modifying or combining existing ones. This concept is often referred to as "Nature is tinkering" (a term coined by Nobel Laureate in Physiology or Medicine, François Jacob, in the 1970s). Algorithmic information theory, which originated in the 1960s, posits that the complexity of an entity can be measured by the length of the shortest algorithm that generates it. Viewing life systems as algorithms and applying algorithmic information theory to these systems began in the 1990s. We have developed a new method that uses algorithmic information theory to explore information with hierarchical nested structures during evolution. This information helps to depict the relationship between the complexity of protein sequences and their functions, quantify the relationship between the structure complexity of neural networks and their performance, design new peptide drugs through this hierarchical nested relationship, compress texts, and even accomplish text classification tasks effectively. 

个人简介：

刘宇，北京师范大学珠海校区文理学院系统科学系副研究员。2018年于瑞典Uppsala大学获得应用数学与统计博士学位，2018-2021年分别在荷兰Groningen大学、瑞典Mittag-Leffler数学研究院、英国Glasgow大学任副研究员。曾在瑞典Mittag-Leffler数学研究院独立主持科研项目。研究方向包括生物系统数学建模、非平衡态系统、群体行为、网络动力学、神经网络、动力学系统(dynamical systems)、随机模拟（stochastic simulation）等。其成果发表在Science Advances, Physical Review Research, Pattern Recognition, Journal of Biological Chemistry等学术期刊。

 

联系人：胡杰