报告时间：12月26日，周四，上午9:00-10:00

报告地点：D533学术报告厅

报告题目：Protein Contact Prediction and Contact-assisted Structural Modeling

报  告  人：龚海鹏博士，清华大学

报告摘要：

Predicting the structures of proteins from amino acid sequences is of great importance. Recently, the accuracy of de novo protein structure prediction has been substantially improved when assisted by the contact information between residues, which is also predictable from sequence. Here we show a novel pipeline for rapid protein structure prediction, which consists of a residue contact predictor AmoebaContact and a contact-assisted folder GDFold. Unlike mainstream contact predictors that utilize simple, regularized neural networks, AmoebaContact adopts a set of network architectures that are optimized for contact prediction through automatic searching and predicts contacts at a series of cutoffs. Different from conventional contact-assisted folders that only use top-scored contact pairs, GDFold considers all residue pairs from the prediction results of AmoebaContact in a differentiable loss function and optimizes atom coordinates using the gradient descent algorithm. Combination of AmoebaContact and GDFold allows quick modeling of the protein structure with acceptable model quality.

报告人简介：

龚海鹏博士，清华大学生命科学学院副教授、博士生导师，生物信息学教育部重点实验室副主任，清华大学结构生物学高精尖创新中心PI。1997年获清华大学生物科学与技术系学士学位；2000年获清华大学生物科学与技术系硕士学位；2006年获Johns Hopkins University生物物理学博士学位；2006-2007年，在Johns Hopkins University从事博士后工作；2007-2009年在University of Chicago大学从事博士后工作，2015年至今担任清华大学生命科学学院副教授。近年来主要从事与蛋白质结构相关的计算方法研究，尤其集中在蛋白质三级结构预测算法研究，以第一作者（\通信作者）在PNAS、Physical Review letters、Nature Machine Intelligence、Journal of Physical Chemistry Letters、Bioinformatics、PLoS Computational Biology等主流期刊发表多篇论文，并提出了RDb₂C、DeepFragLib、AmoebaContact和GDFold等创新性算法。