“机器学习和大数据系统”研讨会系列报告 （6月6日 上午 8:30-12:00）

主持人：李志兵 教授

欢迎广大师生前来参与！

报告一：肿瘤免疫疗法中自然杀伤细胞行为的实验研究    

报告人：史珏 副教授 （香港浸会大学物理系）

【摘要】介绍实验室条件下自然杀伤细胞和肿瘤细胞相互作用的行为观察结果。

Biological processes occur in a wide spectrum of spatial and temporal scales, thus presenting a major challenge to identify the underlying quantitative principles. Although major advancement in molecular genetics, genomics and proteomics in recent years have elucidated the vast number of molecular players that mediate cellular signaling and cell-cell interactions, providing functional maps of the key cellular networks/pathways involved, our understanding of many bio-processes remains limited, in particular regarding how complex network/pathway act dynamically within and between interacting cell types to control physiological responses in complex environments. In this talk, I'll present results from two recent studies in my lab, which attempt to dissect complex cellular dynamics at the system level by combining quantitative single-cell imaging with computational modeling. The first study explored the dose-dependent and cancer cell-type dependent response to DNA damaging drug, one of the most common anticancer chemotherapeutics used in the clinic. We not only uncovered an intriguing bimodal dynamics experimentally that regulate this important drug response process but also pinpointed a four-component molecular module out of a large cellular network that exerts the major control over this bimodal switch. And variable activation of this four-component module accounts for drug response variability in distinct cancer types. The second study investigated the interaction dynamics by which Natural Killer (NK) cells, an important effector cell in the human immune system, detect and kill cancer cells. By imaging unique fluorescent reporter contructs in live cells, we were able to determine the rate-limiting kinetics in NK-cancer cell interaction and identify mechanistically relevant heterogeneity in the process. NK cells are considered promising candidate for cancer treatment, especially for eliminating residual cancer cells after conventional therapy. Results from our current and further study would likely provide new mechanistic insight to develop this cancer immunotherapy.

【个人简介】

香港浸会大学物理系副教授

B.S. in Physics at Zhongshan University, China, PRC

M.A. in Science Education,
Ph.D. in Biophysics at the University of Michigan, Ann Arbor

Postdoctoral in Department of Systems Biology, Harvard Medical School

Website: Center for Quantitative Systems Biology

My main research interest is to combine quantitative measurements with empirical modeling to explore the mechanism of collective behaviors in cellular processes and diseases. In particular, through image-based single cell analysis and kinetic modeling, I would like to characterize and quantify cellular changes induced by anticancer drugs across different cancer types. This would not only improve our understanding of kinetic and molecular alterations in a complex cellular network, induced by local perturbations from drug treatment, but also provide important insight into principles of control and regulation underlying systematic behavior.

The major experimental approach is microscopy, including high-throughput automatic microscopy, and real-time live cell microscopy based on phase-contrast and fluorescence imaging. Other cellular and molecular biology techniques are also employed, such as RNA interference (RNAi), cloning to generate fluorescent reporter cell lines, immunofluorescence, etc. Development of image analysis tools and computational algorithms for kinetic modeling is another essential component of my research interest, as I believe being quantitative is the key to unravel the complexity of bio-systems.

报告二：Monte Carlo  simulations in the analysis of inconsistency indices – recent advancements in the pairwise-comparison-based decision making"（关于分段比较决断不自洽性分析的蒙特卡罗模拟新进展）

报告人：prof. Andrzej Z Grzybowski （琴斯托霍瓦理工大学，机械工程与计算科学学院 数学研究所主任）

【摘要】介绍分析一种大数据比较方法的不自洽性的新近进展。

【个人简介】

Andrzej Z. Grzybowski  graduated  from  the department of Fundamental Problems of Technology at the Wroclaw University of Technology,  Poland,  where he received MSc degree in mathematics. In 1991 he defended his PhD in the field of mathematics at the same university.  In 2013 he completed his habilitation in the field of informatics in the Institute of System Engineering and Informatics, University of Pardubice, Czech Republic.  Since 2016 professor Grzybowski  is a director of  the Institute of Mathematics, Faculty of Mechanical Engineering and Computer Science at the Czestochowa University of Technology. 

His scientific research is devoted to the analysis of various methods in decision theory and their potential applications in industrial engineering problems.  His range   of scientific  interest covers such areas as stochastic control, game theory, evolutionary search methods and its applications, multicriteria decision analysis, regression analysis and some variational problems related to the notion of the Young-measure. Now he is focusing on some artificial intelligence methods, primarily on formal and Monte Carlo analysis of some fundamental problems arising within the multiple criteria decision theory. He is an author and co–author of four scientific monographs and of more than 70 papers published in scientific journals  as well as in a number of World Congresses  proceedings.