报告时间：2026年10月13日10:00-11:00

报告地点：立德楼807

报告名称：Learning Biological Organization from Complex Data

报告摘要：

Biological organization is often not directly observable and needs to be inferred from high-dimensional and incomplete measurements. Recent advances in single-cell and spatial omics technologies have enabled large-scale molecular profiling of biological systems, while also raising new challenges in interpreting, evaluating, and extending these observations.

In this talk, I will present our recent efforts in developing machine learning and statistical approaches for analyzing complex biological data. First, we develop interpretable learning methods to identify molecular factors underlying spatial organization from high-dimensional spatial transcriptomic measurements. Second, we introduce data-driven criteria for evaluating biological structures and selecting appropriate levels of resolution based on the information supported by observed data. Finally, we develop generative modeling approaches to learn spatial molecular distributions from limited observations and enable three-dimensional tissue modeling.

Together, these studies explore how computational methods can help extract, evaluate, and model biological organization from complex data.

专家简介：张驰浩，中国科学院数学与系统科学研究院助理研究员。曾访问美国加州大学洛杉矶分校统计系，并于日本东京大学国际神经智能研究中心从事博士后研究工作。主要研究方向为机器学习与计算生物学，致力于发展面向复杂生物数据的统计学习与智能分析方法。主持国家自然科学基金青年项目，参与国家重点研发计划等科研项目。近年来，以第一作者（含共同第一作者）在 Nature Communications、Nucleic Acids Research、Journal of Machine Learning Research、IEEE Transactions on Pattern Analysis and Machine Intelligence、IEEE Transactions on Knowledge and Data Engineering、Pattern Recognition 等国际期刊发表多篇论文。