The use of orthogonal projection to visualize mosaic domains from topographic data collected on protein crystals

Jeffrey J. Lovelace, Gloria E Borgstahl

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

Owing to the relatively low intensity of diffraction and resulting low contrast topographic images, and also the richly detailed domain structure, it can be difficult to extract individual mosaic domains (size and shape) for protein crystals. Here, orthogonal projection was tested and found to be a superior approach over previous methods to extract mosaic domain information from a fine-sliced topographic sequence of images. The topographic sequence was collected at room temperature from a lysozyme crystal with a low mosaicity. Orthogonal projection can be applied to image sequences that are spatially invariant and composed of linearly additive image formation processes. The domains were determined by using a particle swarm optimizer to fit Gaussians to the integrated intensity profile of the reflection as a function of angle, although any basis function could have been used. This optimization method was more amenable to automation and converged to the global minimum more efficiently. The number of domains can normally be determined by visually inspecting the integrated intensity profile; this only needs to be done once for each crystal, because the number of domains remains constant during collection, and not for each reflection. The results can be used to provide a picture of the internal structure of the crystal and may be useful to aid in the improvement of crystal growth techniques.

Original languageEnglish (US)
Pages (from-to)907-912
Number of pages6
JournalJournal of Applied Crystallography
Volume43
Issue number4
DOIs
StatePublished - Jul 16 2010

Fingerprint

Crystals
Automation
Muramidase
Crystallization
Proteins
Temperature
Image processing
Diffraction

Keywords

  • digital topography
  • eigenvector filtering
  • feature extraction
  • fine slicing
  • mosaic domains
  • orthogonal projection
  • particle swarm optimization

ASJC Scopus subject areas

  • Biochemistry, Genetics and Molecular Biology(all)

Cite this

The use of orthogonal projection to visualize mosaic domains from topographic data collected on protein crystals. / Lovelace, Jeffrey J.; Borgstahl, Gloria E.

In: Journal of Applied Crystallography, Vol. 43, No. 4, 16.07.2010, p. 907-912.

Research output: Contribution to journalArticle

@article{45a01d42d5f541d996fe9e23881cde73,
title = "The use of orthogonal projection to visualize mosaic domains from topographic data collected on protein crystals",
abstract = "Owing to the relatively low intensity of diffraction and resulting low contrast topographic images, and also the richly detailed domain structure, it can be difficult to extract individual mosaic domains (size and shape) for protein crystals. Here, orthogonal projection was tested and found to be a superior approach over previous methods to extract mosaic domain information from a fine-sliced topographic sequence of images. The topographic sequence was collected at room temperature from a lysozyme crystal with a low mosaicity. Orthogonal projection can be applied to image sequences that are spatially invariant and composed of linearly additive image formation processes. The domains were determined by using a particle swarm optimizer to fit Gaussians to the integrated intensity profile of the reflection as a function of angle, although any basis function could have been used. This optimization method was more amenable to automation and converged to the global minimum more efficiently. The number of domains can normally be determined by visually inspecting the integrated intensity profile; this only needs to be done once for each crystal, because the number of domains remains constant during collection, and not for each reflection. The results can be used to provide a picture of the internal structure of the crystal and may be useful to aid in the improvement of crystal growth techniques.",
keywords = "digital topography, eigenvector filtering, feature extraction, fine slicing, mosaic domains, orthogonal projection, particle swarm optimization",
author = "Lovelace, {Jeffrey J.} and Borgstahl, {Gloria E}",
year = "2010",
month = "7",
day = "16",
doi = "10.1107/S0021889810020157",
language = "English (US)",
volume = "43",
pages = "907--912",
journal = "Journal of Applied Crystallography",
issn = "0021-8898",
publisher = "International Union of Crystallography",
number = "4",

}

TY - JOUR

T1 - The use of orthogonal projection to visualize mosaic domains from topographic data collected on protein crystals

AU - Lovelace, Jeffrey J.

AU - Borgstahl, Gloria E

PY - 2010/7/16

Y1 - 2010/7/16

N2 - Owing to the relatively low intensity of diffraction and resulting low contrast topographic images, and also the richly detailed domain structure, it can be difficult to extract individual mosaic domains (size and shape) for protein crystals. Here, orthogonal projection was tested and found to be a superior approach over previous methods to extract mosaic domain information from a fine-sliced topographic sequence of images. The topographic sequence was collected at room temperature from a lysozyme crystal with a low mosaicity. Orthogonal projection can be applied to image sequences that are spatially invariant and composed of linearly additive image formation processes. The domains were determined by using a particle swarm optimizer to fit Gaussians to the integrated intensity profile of the reflection as a function of angle, although any basis function could have been used. This optimization method was more amenable to automation and converged to the global minimum more efficiently. The number of domains can normally be determined by visually inspecting the integrated intensity profile; this only needs to be done once for each crystal, because the number of domains remains constant during collection, and not for each reflection. The results can be used to provide a picture of the internal structure of the crystal and may be useful to aid in the improvement of crystal growth techniques.

AB - Owing to the relatively low intensity of diffraction and resulting low contrast topographic images, and also the richly detailed domain structure, it can be difficult to extract individual mosaic domains (size and shape) for protein crystals. Here, orthogonal projection was tested and found to be a superior approach over previous methods to extract mosaic domain information from a fine-sliced topographic sequence of images. The topographic sequence was collected at room temperature from a lysozyme crystal with a low mosaicity. Orthogonal projection can be applied to image sequences that are spatially invariant and composed of linearly additive image formation processes. The domains were determined by using a particle swarm optimizer to fit Gaussians to the integrated intensity profile of the reflection as a function of angle, although any basis function could have been used. This optimization method was more amenable to automation and converged to the global minimum more efficiently. The number of domains can normally be determined by visually inspecting the integrated intensity profile; this only needs to be done once for each crystal, because the number of domains remains constant during collection, and not for each reflection. The results can be used to provide a picture of the internal structure of the crystal and may be useful to aid in the improvement of crystal growth techniques.

KW - digital topography

KW - eigenvector filtering

KW - feature extraction

KW - fine slicing

KW - mosaic domains

KW - orthogonal projection

KW - particle swarm optimization

UR - http://www.scopus.com/inward/record.url?scp=77954491312&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77954491312&partnerID=8YFLogxK

U2 - 10.1107/S0021889810020157

DO - 10.1107/S0021889810020157

M3 - Article

AN - SCOPUS:77954491312

VL - 43

SP - 907

EP - 912

JO - Journal of Applied Crystallography

JF - Journal of Applied Crystallography

SN - 0021-8898

IS - 4

ER -