Albeit the number of globular proteins structures being determined is rapidly increasing, the determination of membrane protein structures via experimental approaches remains a major challenge in the field of structural genomics. As a rule, it is also more costly to determine the structures of membrane proteins experimentally. Thus, computer modeling may be served as a viable way to obtain membrane protein structures prior to experimental methods. The human adenosine A_2A receptor has recently been identified as a candidate target for designing therapeutics for the Huntington disease. We use ab initio approach to construct full length A_2A receptor model and refine it by molecular dynamics simulations. Finally, docking simulations were conducted to predict the binding sites of known substrates and inhibitors to validate the model.

(02)23123456-8404

Some small drug molecules are able to interact with DNA and are often used as anticancer drugs, such as doxorubicin, daunomycin, mitoxantrone, etc. The details of drug mechanisms are still under investigation, but it is generally thought to inhibit the replication, transcription of DNA, or the activity of topoisomerases. To know the details how these drugs work, it is necessary to understand the interaction between DNA and small drug molecules. There are three types of non-covalent interactions between DNA and small drug molecule: electrostatic binding, groove-binding, and intercalation. And my research mainly focuses on using molecular modeling to investigate the intercalation process between DNA and drug molecules.

I am interesting in protein-ligand binding free energy calculation and prediction.
I try to use molecular dynamics methods to improve the prediction of protein-ligand binding free energy rather than that in the static structure of proteins.

I am interested in protein bioinformatics. In the past, I have investigated in protein structural alignment during my master period. The study involved local structures analysis, clustering, superimposition and alignment algorithm. My current research topics include: investigation of neuraminidase from sequence mining and phylogenesis to molecular dynamics simulation for drug design; proteins and ligands database.

Learning and researching the microscopic world is my interesting. The computation method and analysis for F₁-ATP mechanism are the focus of my immediate researching. ATP is the primary matter which supports energy to the biological cells. According to the computer advance, we can use molecular dynamics simulation method to understand the conformations and some details of the ATP protein. With this work, we can understand the ion channel and make some suggestion about remedy for related disease

My research is to analyze the data about pharmacy with Statistic Methods. I use regression analysis and ANOVA table to find the best linear model and the confidence interval, and then test the correlation between the data.
In the future, we might extend our analysis to nonlinear model.

I am interested in applying high performance computing on molecular dynamic calculations. Through 64bit parallel computing of PC cluster system, we can tackle much bulkier molecular systems of proteins, DNA or lipid simulations and analyze their thermodynamic characteristics. I am also interested in prediction of protein structures. Predict protein secondary structure by machine learning algorithms and tertiary structure by threading methods. We try to use computing or simulation methods to unveil the mysteries of protein structures.

My research interest is expolring statistical mechanics of the biomolecule by computer simulation. The simulated method is molocular dynamics and the related software is Amber.

My research interest is using 3D-pharmacophore model to predict molecular binding constant and use the pharmacophore model to discuss the binding mechanism. This method is often used in the protein which is not easily crystallized. 3D-pharmacophore also can use to screen potent moleculars in the molecular database

The prediction of drug properties and bioavailability are confirmed for good drug design. ADME/T properties of drugs are associated with the efficacy of therapeutics. A major problem of drug delivery is multidrug resistance (MDR) phenomenon. The effect of MDR pumps characterized by the extrusion of cytotoxic drug molecules is a kind of protection against the toxic materials.

My research topics include:
1. Multidrug resistance efflux transporters
The potency of antibiotics or chemotherapy decreases owing to overexpression of MDR transporters. Most of these pumps are belonged to membrane protein. And the crystallization of integral membrane protein is a challenge because of their amphipathic nature. Only few high-resolution structures of membrane proteins have been determined. Thus, we investigate the probable process of pumping the substrates by computational methodology to improve the therapy efficacy.

2. Prediction the effects of drug delivery and drug properties with QSAR and QSPR Quantitative structure-activity relationship (QSAR) analysis and quantitative structure-property relationship (QSPR) analysis are major applications in computer-assisted molecular design (CAMD). The above two applications build up the linear models to predict the drug activity or the drug ADME/T properties. The model construction can be used as general utilities to screen the suitable compounds for drug design in a virtual screen fashion.

1. E. coli thioesterase/protease I (TEP-I): TEP-I belongs to a newly discovered subclass of lipolytic enzymes of the serine protease superfamily. It is a versatile enzyme desirable for industrial application for synthesis of stereospecific esters, acids and alcohols. We have determined the dynamics of TEP-I and are currently investigating the role of active site hydrogen bond network in catalytic process.

2. Human branched-chain α-ketoacid dehydrogenase (BCKD): BCKD is a multienzyme complex, catalyzing the oxidative decarboxylation of branched-chain α-ketoacids. Deficiency in BCKD complex causes maple syrup urine disease (MSUD) with a clinical consequence including often-fatal acidosis, neurological derangement and mental retardation

anderson@jlin.mc.ntu.edu.tw

fiesta1221@yahoo.com.tw

http://jlin.mc.ntu.edu.tw/~anderson

• Ligands and proteins complexes' experimental ki values data mining and managing.
• PDB data base construction and managing.
• Scoring function(s) design and investigation of new molecular descriptors.
• Class optimized scoring function(s) design.
• AutoDock scoring function and AutoGrid inspection.
• Ligands and proteins data processing and managing.
• Thermodynamic Integration and automation.

patricia@jlin.mc.ntu.edu.tw

http://rx.mc.ntu.edu.tw/~patricia/

Researches Interests

• Grid
• Bioinformatics
• PC Cluster
• Web Services
• Microarray analysis