|
报告时间：11月22日（星期日）8:00 报告形式：腾讯会议（会议ID：170 156 944，密码：202022） 报 告 人：黄德双，同济大学教授 报告题目：Computational Analyses for Transcription/Translation Factors Binding Data 内容简介：Transcription/Translation factor (TF) play a central role in gene regulation. Knowing the binding specificities of TFs is essential for developing models of the regulatory processes in biological systems and for deciphering the mechanism of gene expression. In this talk, we will introduce several novel computational models of TF binding data by combining various types of high-throughput data. Firstly, we will introduce a tensor decomposition model for collaborative prediction of ChIP-seq data, which could overcome its current limitation for integrative analysis. Secondly, we will present a de novo motif learning method based on the area under the receiver-operating characteristic curve (AUC) criterion, which has been widely used in the literature to evaluate the significance of extracted motifs. Finally, based on Fisher Exact Test score (FETS), we propose DirectFS, which is (to our best knowledge) the first FETS-based approach that allows direct learning of the motif parameters in continuous space. Experimental results based on real world high-throughput datasets illustrate that DirectFS outperforms competing methods for refining motifs found by de novo motif elicitation methods, while being one order of magnitude faster. 报告人简介：黄德双，同济大学计算机科学系教授，博士生导师，机器学习与系统生物学研究所所长。国际模式识别协会Fellow (IAPR Fellow), IEEE和国际神经网络学会INNS的高级成员, IEEE计算智能学会生物信息学与生物工程技术委员会以及神经网络技术委员会委员。任IEEE/ACM Transactions on Computational Biology and Bioinformatics 及Neural Networks的副主编。国际智能计算学术会议（ICIC）发起人、大会主席并担任每年举办的ICIC会议指导委员会主席。此外，担任2015年神经网络国际联席会议(IJCNN 2015，爱尔兰)总主席，以及2014年第11届IEEE生物信息学和计算生物学计算智能会议(IEEE-CIBCBC，美国)委员会主席。主要研究兴趣包括神经网络、模式识别和生物信息学等。 |
