|
Activity Number:
|
68
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, July 29, 2007 : 4:00 PM to 5:50 PM
|
|
Sponsor:
|
Section on Statistical Computing
|
| Abstract - #309637 |
|
Title:
|
Calculating the Interatomic Distance Distribution from Small-Angle X-Ray Scattering via Curve Averaging
|
|
Author(s):
|
Lanqing Hua*+ and Alan Friedman and Chris Bailey-Kellogg and Bruce Craig
|
|
Companies:
|
Purdue University and Purdue University and Dartmouth University and Purdue University
|
|
Address:
|
2108 W. White Street, Champaign, IL, 61821,
|
|
Keywords:
|
indirect transform ; model averaging
|
|
Abstract:
|
Small-angle X-ray scattering (SAXS) is a useful technology to study a molecule's structure. Relative to other technologies, it provides structural information with a reduced investment in time and effort. Reconstruction of the interatomic distance distribution, P(r) curve, has been a long standing problem with SAXS experiments. In this paper, we present an indirect transform method using "model averaging". This approach expands on Moore's (1980) method by using multiple sets of basis functions to fit the observed scattering data and reconstructing the P(r) curve as a weighted average of these different representations. This approach completely avoids the need to impose subjective physical assumptions about the protein or restraints on P(r), common with other approaches. Through simulation, we show that our method provides comparable or more accurate P(r) estimation than Moore's method.
|
