黃鎮剛 計算生物學教授 國立交通大學 生物資訊研究所
Jenn-Kang Hwang, Professor of Computational Biology
Institute of Bioinformatics
National Chiao Tung University
HsinChu, Taiwan
One generation passeth away, and another generation cometh: but the earth abideth for ever. Ecc 1: 4
 
  research.html

 黃鎮剛 教授
 國立交通大學 生物科技學院院長
 生物資訊研究所

 Professor Jenn-Kang Hwang
 Dean, College of Biol. Sci. & Tech.
 Institute of Bioinformatics
 National Chiao Tung University, HsinChu, Taiwan

 E-mail: jkhwang (at) cc . nctu . edu . tw
 Tel : +886-3-513-1337
 Fax: +886-572-9288

 Research Topics
  • Structure prediction and classification
  • Prediction of protein subcellular localization
  • Disulfide proteins
  • Protein stability
  • Structural alignment
  • Molecular Simulation
  • Sequence coupling
  • Whatever is fun
 Academic activities


Radio Plot of B-factors (red) and the sequare of centroid distance (black) of 1U0S

Protein subcellular localization  
Due to recent progress in genomics research, incredible amounts of DNA and protein sequence data are generated in great speed from experiment. Hence, to develop useful computational tools to extract relevant biological information from sequences becomes even more important nowadays. Due to the close correlation between the subcellular location of a protein and its biological function, the knowledge of subcellular location can help to characterize the protein functions. Therefore, a reliable, automated prediction tool for subcellular localization will be useful for the characterization of expressed protein sequences. We have developed a method CELLO based on Sopport Vector Machine to predict the subcellular locations for both eukaryotes and prokaryotes.

See Nature Review Microbiology for a review of CELLO
Structural entropy and thermostability
We developed a technique to compute the structural entropy directly from protein sequences. We explored the possibility of using structural entropy to identify residues involved in thermal stabilization of various protein families. We observed a good linear relationship between the average structural entropy and the melting temperatures for adenylate kinase and its chimeric constructs.

Structural alignment  
Identify functional structural motifs from structures of unknown function becomes increasingly important in recent years due to the progress of the structural genomics projects. Though some structural patterns such as the Asp-His-Ser catalytic triad are easy to detect because of their conserved residues and the stringently constrained geometry, it is usually more challenging to detect a general structural motifs like, for example, ββα-metal binding motif, which has a much more variable conformation and sequence. At present, the identification of these motifs usually replies on manual procedures based on different structure and sequence analysis tools. We have developed a structural alignment algorithm combining both structural and sequence information to identify the local structure motifs.
Molecular Simulation
Understanding the molecular origin of enzyme catalysis is one of the most fundamental problems in molecular biology. The effects of enzymes on the reacting substrates involve varied and complicated energy contributions which depend on the structure of the enzyme-substrate complex. Thus to have a quantitative structure-function relationship is not a trivial matter. In principle, one would like to carry out ab initio quantum mechanical calculations for biomolecules. Unfortunately, such an approach cannot be implemented at present in studies of proteins.
One must therefore resort to some approximations and represent a large part of the system classically. The enzymatic reactions are described by the empirical valence bond (EVB) method, which is probably one of the most practical and consistent approaches for treating chemical reactions both in enzymes and in solution. The basic idea is to represent the overall reaction in terms of various valence bond structures and then to obtain the actual ground state by mixing these structures. The EVB potential surface is calibrated by using accurate gas-phase data or by using experimental information for the corresponding reactions in solution.
QM/MM simulation using Feynman's quantum beads


In combination with the free energy perturbation (FEP) method, the EVB-FEP method offers a convenient way to calculate the free energy profile for the enzymatic reactions. We have carried out many EVB-FEP calculations on enzymes such as serine proteinases, aspartate aminotransferase and human carbonic anhydrase I. Recently, we have proposed the quantized classical path (QCP) approach which incorporates the Feynman's path integral into the EVB framework, and offers a convenient way to obtain quantum-mechanical corrections to the rate constants of proton transfer (PT) and hydride transfer reactions in enzymes and in solution. Efforts are also invested in combining the semi-empirical quantum mechanical method such as AM1 with the EVB- FEP approach.

There is an interesting article in Nature reporting the recent progress in the study of QM effects on enzyme reactions and our related work.
 
Links  
The Molecular Bioinformatics Center
CELLO
(PS)2
RCSB PDB
SCOP
NCBI
The Structural Genomics Consortium
Protein interaction domains
nrpdb
PDP

Nature
Science
Nature: Structural & Molecular Biology
Nature: Biotechnology
Nature: Medicine
Structure
Current Opinion in Structural Biology
Cell
EMBO
PNAS
Genome Research
Nucleic Acids Research
Journal of Molecular Biology
Protein Science
Proteins: Structure, Function, and Bioinformatics
Journal of American Chemical Society
Accounts of Chemical Research
Recent Patents on Anti-Infective Drug Discovery

COMICS and OTHER FUN STUFF

Rice University
Anderson
Auth
Dilbert
Carlson
Oliphant
Bok
Sargent
Doonsbury
Garfield
Bob's songs
Poe
Maimonides
Chinese etymology
Art Renewal
Kitvei HaKodesh
Jewish Humor
典籍 (Chinese)
韓非子 (Chinese)
慎子 (Chinese)
夢溪筆談 (Chinese)
西遊記 (Chinese)
三國演義 (Chinese)
颮翭 (Chinese)

Publications
Links