Ellipsoid-weighted protein conformation alignment

Hyuntae Na, Guang Song

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Conformation alignment is a critical step for properly interpreting protein motions and conformational changes. The most widely used approach for superposing two conformations is by minimizing their root mean square distance (RMSD). In this work, we treat the alignment problem from a novel energy-minimization perspective. To this end we associate each atom in the protein with a mean-field potential well, whose shape, ellipsoidal in general, is to be inferred from the observed or computed fluctuations of that atom around its mean position. The scales and directions of the fluctuations can be obtained experimentally from anisotropic B-factors for crystal structures or computationally. We then show that this "ellipsoid-weighted" RMSD alignment can be reformulated nicely as a point-to-plane matching problem studied in computational geometry. This new alignment method is a generalization of standard RMSD and Gaussian-weighted RMSD alignment. It is heavily weighted by immobile regions and immobile directions of the protein and hence highlights the directional motions of the flexible parts. It has an additional advantage of aligning conformations of proteins along their preferred directions of motions and could be applied to order protein conformations along its trajectory.

Original languageEnglish (US)
Title of host publicationBioinformatics Research and Applications - 9th International Symposium, ISBRA 2013, Proceedings
Pages273-285
Number of pages13
DOIs
StatePublished - Sep 9 2013
Event9th International Symposium on Bioinformatics Research and Applications, ISBRA 2013 - Charlotte, NC, United States
Duration: May 20 2013May 22 2013

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume7875 LNBI
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

Other9th International Symposium on Bioinformatics Research and Applications, ISBRA 2013
CountryUnited States
CityCharlotte, NC
Period5/20/135/22/13

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Fingerprint Dive into the research topics of 'Ellipsoid-weighted protein conformation alignment'. Together they form a unique fingerprint.

  • Cite this

    Na, H., & Song, G. (2013). Ellipsoid-weighted protein conformation alignment. In Bioinformatics Research and Applications - 9th International Symposium, ISBRA 2013, Proceedings (pp. 273-285). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 7875 LNBI). https://doi.org/10.1007/978-3-642-38036-5_27