1993 Fiscal Year Final Research Report Summary
The Method of Automatically Building Tertiary Structure of Proteins from Their Primary Structure
| Project/Area Number |
04455007
|
|---|
| Research Category |
Grant-in-Aid for General Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Research Field |
広領域
|
|---|
| Research Institution | University of Tokyo |
|---|
Principal Investigator |
NAGANO Kozo Univ.Tokyo, Fac.Pharm.Sci., Prof., 薬学部, 教授 (30012636)
|
|---|
| Project Period (FY) |
1992 – 1993
|
| Keywords | protein / tertiary structure / amino acid sequence / supersecondary structure / alpha / beta proteins / domain / atomic coordinates / algorithm |
|---|
| Research Abstract |
Tertiary structure of protein is determined from its amino acid sequence. Since the folding process is finished in some mili-seconds, it must not be like searching the most stable conformation among the all geometrically possible conformations, but could be such as a way of combining some limited numbers of candidates having local conformation with each other and finding the most suitable harmonized conformation as a unique representative. Inspection of various tertiary structures obtained so far by X-ray crystallography has shown that some supersecondary structures can be formed in spite of low amino acid sequence homology. Supersecondary structure is a mode of similar packing arrangements of secondary structure such as alpha-helix, beta-strand and turn. An algorithm has been developed by the present head investigator (Nagano) for predicting alpha-helices and beta-strands under the assumption of having alpha/beta barrel proteins and fitting the primary structure to the atomic coordinates accumulated in the Protein Data Bank. If the fitting does not work well, the other types of supersecondary structures can be tried. Simulation has been tried for finding the true patterns of conformation for more than 20 kinds of alpha/beta proteins at the highest rank. When it is obtained at the highest rank, the atomic coordinates of the domain can be calculated immediately by using the partial coordinates stored in the Protein Data Bank and for showing the three-dimensional structure on the screen of a computer graphics display. Such an algorithm has been developed for years. In order to find the true pattern at the highest rank, the mode of interactions between beta-strands, or between alpha-helices and beta-strands, or between alpha-helices has been investigated carefully and improved step by step.
|
