SLAI Seminar 25th Session will be discussing the topic on "Structure Prediction from Proteins to RNAs: Going beyond AlphaFold", from 10:00am to 11:30am, January 26th (Monday) at B311 Lecture Hall, online participation is welcome (Tencent Meeting ID: 245-394-752)

报告主题：从蛋白质到RNA的结构预测：超越AlphaFold

时间：2026年1月26日（周一）上午10:00-11:30

地点: 深圳河套学院B311阶梯教室

线上参与：腾讯会议号245-394-752

讲者简介 About the Speaker:

周耀旗教授现任深圳湾实验室系统与物理生物学研究所资深研究员、副所长，同时是砺博生物科技有限公司的科学创始人，著有畅销书《出发，不断走出舒适区的科研生活之旅》。他曾担任美国印第安纳大学终身正教授及澳大利亚格里菲斯大学讲席教授。

周教授长期致力于结构生物信息学研究，早在2009年率先将浅层神经网络学习应用于连续蛋白质骨架二面角预测，为端到端蛋白质结构预测奠定基础，并预见了后来获得诺贝尔奖的AlphaFold 2技术雏形。2014年，他主导开发了人工智能驱动的蛋白质序列设计方法SPIN/SPIN2，被业界誉为"人工智能应用于蛋白质设计的起点"。周教授研究团队在历届国际蛋白质/RNA结构预测与功能预测竞赛中名列前茅。周教授的谷歌学术引用量逾2万次，H指数达79。

归国后，周教授先后承担科技部、国家自然科学基金及广东省科技厅重大科研项目。周教授目前聚焦于基于人工智能与高通量实验的蛋白质/RNA基础与应用研究，以及药物研发与递送系统的创新。

2022年，他联合创办砺博生物科技有限公司，打造国际领先的AI驱动、干湿实验闭环整合的RNA靶点与小分子药物发现平台，致力于开发突破性RNA靶向疗法，应对未满足的临床需求，该公司近期已完成近亿元人民币的Pre-A轮融资。

Professor Yaoqi Zhou is a Senior Investigator at Shenzhen Bay Laboratory and Associate Director of the Institute of Systems and Physical Biology. He is also the Scientific Founder of Ribopeutic, Inc. and the author of the bestselling book “Departure: A Journey of Scientific Life Beyond Comfort Zones” 《出发，不断走出舒适区的科研生活之旅》. Prior to this, he served as a Full Professor at Indiana University and a Chair Professor at Griffith University in Australia.

With long-term dedication to structural bioinformatics, Professor Zhou pioneered the use of shallow neural network learning in 2009 to predict continuous protein backbone dihedral angles, paving the way for end-to-end protein structure prediction and foreshadowing the Nobel Prize–winning work of AlphaFold 2. In 2014, he led the development of the AI-driven protein sequence design methods SPIN/SPIN2, regarded as "the starting point of artificial intelligence in protein design". His research group has consistently ranked among the top performers in international competitions for protein/RNA structure prediction and functionprediction. His Google Scholar citations exceed 20,000 with an H-index of 79. Since returning to China, he has secured major research grants from the Ministry of Science and Technology, the National Natural Science Foundation of China, and the Department of Science and Technology of Guangdong Province. He currently focuses on AI- and high-throughput experiment–based fundamental and applied research on proteins/RNA, as well as drug development and delivery.

In 2022, he co-founded Ribopeutic, Inc., which developed an internationally leading AI-enabled, dry-wet integrated closed-loop platform for RNA target and small-molecule drug discovery, providing innovative RNA-targeting therapeutics to address unmet clinical needs. The company recently completed nearly 100 million RMB in pre-A funding.

报告摘要 Abstract：

AlphaFold 2 彻底革新了蛋白质结构预测的准确性。然而，其性能在很大程度上依赖于从蛋白质序列库中为给定序列检测到的天然同源物的质量和数量，而这些因素往往难以控制。高质量同源物的缺乏是关键原因之一，导致仅有36%的人类蛋白质组残基能以高置信度被预测。这一问题并未随着AlphaFold 3或其他更新技术的出现而得到解决。对于RNA而言，情况更为严峻：由于RNA在序列空间中的保守性极差，若其二级结构未知，目前根本没有可靠的方法搜索RNA同源物。本研究将展示实验室生成的同源序列与语言模型如何破解这一难题。该进展通过将人工智能与高通量测序技术相结合，为快速、经济高效的蛋白质与RNA结构预测开辟了前景广阔的发展道路。

AlphaFold 2 revolutionized the accuracy of protein structure prediction. However, its performance heavily relies on uncontrollable quality and quantity of natural homologs that can be detected in protein sequence libraries for a given sequence. The lack of quality homologs is one key reason why only 36% of human proteome residues were predicted with high confidence. This problem was not solved with the arrival of AlphaFold 3 or other updated techniques. The situation for RNAs is even worse: there is simply no reliable way of searching RNA homologs if its secondary structure is unknown because RNAs are poorly conserved in sequence space. Here we will show how lab-generated homologous sequences and language models can help. The advancement paves the way for a promising future of rapid, cost-effective structure prediction for proteins and RNAs by integrating AI with high-throughput sequencing.