2006 Fiscal Year Final Research Report Summary
Geometry optimizations of proteins using the ab initio fragment molecular orbital method
| Project/Area Number |
16350017
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Research Field |
Physical chemistry
|
|---|
| Research Institution | National Institute of Advanced Industrial Science and Technology (AIST) |
|---|
Principal Investigator |
KITAURA Kazuo National Institute of Advanced Industrial Science and Technolog, Reserch Institute for Computational Sciences, Principal Reserch Scientist, 計算科学研究部門, 上席研究員 (30132723)
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
UEBAYASI Masami National Institute of Advanced Industrial Science and Technolog, Institute for Biological Resources and Functions, Reserch Scientist, 生物機能工学部門, 主任研究員 (70356559)
NEMOTO Tadasi National Institute of Advanced Industrial Science and Technolog, Biological Information Research Center, Reserch Scientist, 生物情報解析センター, 主任研究員 (70357739)
|
|---|
| Project Period (FY) |
2004 – 2006
|
| Keywords | fragment MO method / protein / geometry optimization / ab initio molecular orbital / FMO method |
|---|
| Research Abstract |
We have developed the fragment molecular orbital method (FMO) for the electronic structure calculations of large molecules such as proteins. In the FMO method, a molecule is divided into small fragments and ab initio MO calculations are performed on the fragment monomers and their dimers to obtain the total energy and other properties of while molecule. In this work, we have developed the analytic derivative of the electrostatic interaction between far separated fragments necessary for practical gradient calculations of large systems, and performed a number of geometry optimizations using FMO and ab initio methods. In particular, the α-helix, β-turn and extended conformers of 10-residue polyalanine were studied, and the good FMO accuracy established (RMSD for the former two forms were about 0.2 Å and for the latter structure 10^<-3> Å). Met-enkephalin dimer used as a model for polypeptide binding was computed at the 3-21G and 6-31G^* levels and a similar accuracy was achieved; the error in the binding energy predictions (FMO vs ab initio) was 1-3 kcal/mol. Chignolin (PDB : luao) and an agonist polypeptide of erythropoietin receptor protein (emp1) were optimized at the 3-21(+)G level, with the RMS deviation from ab initio of about 0.2 Å, or 0.5^O° in terms of bond angles. It is shown that the FMO method can be practically used for the geometry optimization of real proteins.
|
