1. 报告题目: Pathological Image Classification Based on Deep Learning Methods

报告人：张法 中国科学院计算技术研究所 教授

报告时间：2019年12月21日 8:30-9:30

报告地点：JC1011

报告摘要：

Histopathological diagnosis is considered the gold standard in diagnosing cancer. Although the application of deep learning has greatly improved the performance of Pathological Image Classification, the accuracy of classification has been unable to meet the requirements of clinical practice. In this talk, I will introduce two kind of breast cancer histopathological image classification based on deep learning methods, one is a hybrid convolutional and recurrent deep neural network for breast cancer histopathological image classification, which integrates the advantages of convolutional and recurrent neural networks. The other is a new hybrid deep learning method for breast cancer classification using fusion of multimodal data.

报告人简介：

张法，博士，中国科学院计算技术研究所研究员，博士生导师。主要从事生物信息学算法和高性能计算方面的研究，近年来在冷冻电镜三维重构、生物医学图像处理等方面，取得了多项重要研究成果。在Cell Research、Bioinformatics和Journal of Structural Biology等国际著名期刊和ISMB等知名国际会议发表论文100余篇。作为项目负责人和主要参与人承担了多项科技部重点研发专项、国家自然科学基金重点和国际合作重大项目、中科院先导、中科院知识创新重点等项目。担任中国计算机学会生物信息学专业委员会秘书长，中国生物物理学会冷冻电镜分会理事，中国人工智能学会生物信息学与人工生命专委会常委。

2. 报告题目：Finding Remote Homologous Proteins: Alignment-Based, Alignment-Free and Cross-Modal Methods

报告人：崔学峰 山东大学计算机科学与技术学院 教授

报告时间：2019年12月21日 9:30-10:30

报告地点：JC1011

报告摘要：

Proteins function in living organisms as enzymes, antibodies, sensors, and transporters, among myriad other roles. Understanding protein functions has great implications for the study of biological and medical sciences. Finding remote homologous proteins, with conserved structure similarities but limited sequence similarities, is an indispensable step towards understanding protein functions.

Here, three novel methods are presented for finding remote homologous proteins with different goals: (a) the PROtein STructure Alignment (PROSTA) methods that automatically determine and align homologous structures of protein pockets and interaction interfaces; (b) the ContactLib method that scans tens of thousands of protein structures for homologous structures in seconds; (c) the CMsearch method that simultaneously explore the sequence space and the structure space to perform cross-modal search for homologous proteins. Our methods do not only improve the accuracy of finding homologous proteins, but also improve the accuracy of predicting protein structures. Moreover, case studies where our method discovers, for the first time, structural similarities between pairs of functionally related protein-DNA complexes are presented。

报告人简介：

崔学峰，现任山东大学计算机科学与技术学院教授。在加拿大滑铁卢大学先后获得了本科、硕士、博士学位。硕士与博士导师为加拿大基拉姆奖（Killam Prize，加拿大最高科研奖）得主、加拿大皇家科学院院士、ACM会士、IEEE会士 李明 讲席教授（University Professor）。2014年博士毕业后，在沙特阿拉伯阿卜杜拉国王科技大学（KAUST）担任了两年多的博士后。2016年回国后，在清华大学交叉信息研究院担任了三年的Tenure-Track助理教授。主要科研领域为生物信息学。一直致力于设计机器学习与并行算法，用来解决与人类生活息息相关的生物问题。第一作者论文3次发表在会议Intelligent Systems for Molecular Biology（ISMB，生物信息学顶级会议，每年仅录取约40篇论文）。此外，创新科研成果被国际媒体Bio-Techniques报道1次，被国际媒体Science X报道2次。

3. 报告题目：Cancer Subtype Analysis based on Multi-Omics Data

报告人：郑春厚 安徽大学计算机科学与技术学院 教授

报告时间：2019年12月21日 10:30-11:30

报告地点：JC1011

报告摘要：

Cancer is one of the main diseases threatening the safety of human life. With the development of high-throughput sequencing technology, a large amount of multi-omics bio-molecular data has been generated, which brings opportunities to the research on the mechanism and therapy of cancer. A series of computational methods have been proposed to make effective use of the data. Among them, the discovery of cancer subtypes has become one of the research hotspots in oncology and bioinformatics. Dividing cancer patients into different subtypes can provide basis and guidance for precision medicine and personalized medicine, so as to improve the treatment effect, as well as provide assistance for cancer mechanism analysis and drug target research and development. This talk will give out our research work about cancer subtype analysis based on Multi-Omics Data, including Scluster, ndmaSNF (network diffusion model assisted SNF) and RWCE (Random Walk based Cluster Ensemble).

报告人简介：

郑春厚，安徽大学计算机科学与技术学院教授、博士生导师，安徽省学术和技术带头人后备人选。近年来，在Bioinformatics、Neural Computation、Pattern Recognition、IEEE/ACM Transactions 系列会刊等国内外重要学术刊物与国际会议上发表论文100余篇，论文总被引2000余次。主持国家自然科学基金项目4项、省部级课题多项。2007年获中国科学院王宽诚博士后工作奖，2010年获安徽省自然科学一等奖（第二完成人），2016年获教育部自然科学一等奖（第二完成人）。近年来，应邀在多个国际、国内学术会议做交流报告。多次担任国际学术会议“International Conference on Intelligent Computing” Publicity Co-Chair及Program Committee Member。现任中国计算机学会生物信息学专业委员会委员、中国自动化学会人工智能与机器人教育专业委员会委员。