Feng Zhao

About Me
I obtained my Ph.D on Computer Science at Toyota Technological Institute at Chicago under the guidance of Professor Jinbo Xu on solving the ab intio protein folding problems from 2007 to 2013. I received my B.Sc. degree from Peking University and my MMath degree from University of Waterloo both in Computer Science.

• RaptorX-Roll was invited for the FM roundtable CASP10 meeting in 2012 (Critical Assessment of Techniques for Protein Structure Prediction, a worldwide bi-annual computational modeling competition since 1994)
• RaptorX ranked 2nd place among all server groups and 1st place in human TBM category in CASP9 in 2010
• RAPTOR ranked 2nd place among all servers in the CASP8 competition in 2008.

• Rewrote the pricing and portfolio analysis system (backend engine) to SaaS (Software as a Service) on Infiniband high performance computing environment, improving the speed for 100 times.
• Built risk management toolkits including fast Marginal PML curve calculator, Data Validator, and portfolio optimizer ( a LP solution to the original non-linear non-concave TVaR-constraint problem).

• Built market making strategies using machine learning methods on Treasure Notes
• Built optimal execution strategies for Pro rata contracts.

• Developed a new probabilistic model that is 100 times faster than current NN models. It is applied to analyze over 100 million entries of data with about 3000 dimensions using 1200 cores on super-computers.
• Implemented a novel Conditional Neural Fields model combining the advantages of graphical model and neural network to model nonlinearity on structured data, successfully applied to the protein folding problem.
• Developed a novel sampling algorithm applying the Conditional Radom Fields model with Simulated Annealing MCMC method, which greatly reduced the conformational search space.

• Natural Language Processing supported multi-language Search Engine;
• Single Sign On enabled web-based Enterprise Information System.
• High-end client side imaging processing application on PC and handheld devices;

• EEG&EMG product

1. Protein threading using context-specific alignment potential .
   Jianzhu Ma, Sheng Wang, Feng Zhao, and Jinbo Xu.
   ISMB 2013
   
   
2. A position-specific distance-dependent statistical potential for protein structure and functional study.
   Feng Zhao, Jinbo Xu.
   Structure(Cell Press) 2012, Volume 20, Issue 6.
   Software Download      User Guide      Non-redundant database
   
   
3. RaptorX: low-homology protein threading and multiple-template threading.
   Jinbo Xu, Jian Peng and Feng Zhao.
   CASP9 meeting 2010 Web Server .
   
   
4. Protein 8-class Secondary Structure Prediction Using Conditional Neural Fields.
   Zhiyong Wang, Feng Zhao, Jian Peng, and Jinbo Xu.
   Proteomics, 2011. (An early version appears at BIBM 2010).
   
   
5. Fragment-free Approach to Protein Folding Using Conditional Neural Fields
   Feng Zhao, Jian Peng and Jinbo Xu
   The Eighteenth Annual International Conference on Intelligent System for Molecular Biology (ISMB2010)
   Software Download     User Guide
   
   
6. Efficient Fragment-free Approach to Protein Folding.
   Feng Zhao, Jian Peng and Jinbo Xu.
   The Eighteenth Annual International Conference on Intelligent System for Molecular Biology (ISMB2010), poster.
   
   
7. A probabilistic and continuous model of protein conformational space for template-free modeling.
   Feng Zhao, Jian Peng, Joe DeBartolo, Karl F. Freed, Tobin R. Sosnick and Jinbo Xu.
   Journal of Computational Biology, 2010. Highlight article of Volume 17(6). An early version appears at RECOMB 2009.
   
   
8. A Probabilistic Graphical Model for Ab Initio Folding
   Feng Zhao*, Jian Peng*, Joe DeBartolo, Karl F. Freed, Tobin R. Sosnick and Jinbo Xu
   (*: The first two authors contribute equally to the paper.)
   The 13th Int'l Conference on Research in Comp. Mol. Biol (RECOMB 2009)
   
   
9. Template-based and free modeling by RAPTOR++ in CASP8
   Jinbo Xu, Jian Peng and Feng Zhao
   The CASP8 special issue of PROTEINS. 2009 (invited paper)
   
   
10. Structure prediction combining the template-based RAPTOR algorithm with the ItFix ab initio method
    J. DeBartolo, G. Hockey, F. Zhao, J. Peng, A. Augustyn, A. Adhikar, J. Xu, K. F. Freed and T. R. Sosnick
    Critical Assessment of Techniques for Protein Structure Prediction (CASP8) Meeting, 2008
    (Best Poster Award)
    
    
11. RAPTOR: protein structure prediction by multiple techniques
    Jinbo Xu, Jian Peng and Feng Zhao
    Critical Assessment of Techniques for Protein Structure Prediction (CASP8) Meeting, 2008 (Poster)
    
    
12. Discriminative Learning for Protein Conformation Sampling
    Feng Zhao, ShuaiCheng Li, Beckett W. Sterner and Jinbo Xu.
    PROTEINS: Structure, Function and Bioinformatics, 2008.
    
    

Fan and Player of Table Tennis:
Champion of class F in Schaumburg January Open JanuaryOpenResults-2010 at Schaumburg Table Tennis Club