In recent years it has been found that, under properly controlled conditions, electron diffraction intensity data from the microcrystals of biomolecules, such as lipids, polypeptides and polysaccharides, can be used for ab initio quantitative crystal structure analyses. For lipids, epitaxial orientation produces the most useful projection onto the molecular axis whereas, for linear polymers, both solution and epitaxial crystallization are useful. Initial structure determinations have been carried out using traditional phasing procedures including interpretation of a Patterson map and construction of molecular conformational models suggested from similar crystal structures. We have found that direct phasing techniques based on the use of structure invariant relationships can also be used for such analyses (e.g. for phospholipids, paraffins, linear polymers, and various small organic molecules), particularly if they are combined with phase information derived from high resolution electron microscope images. This frees the structure determination from biases imposed by a model which can be serious when the number of observed diffraction data is relatively small. In this project we would like to extend the development of direct phasing methods for analysis of available lamellar data from a number of yet uncharacterized phospholipids, as well as to three-dimensional data from long chain wax esters (in anticipation of the successful three-dimensional crystallization of glycerolipids) and for two- and three-dimensional data from solution-crystallized biopolymers such as polypeptides and polysaccharides. In some cases, data from non-uniformly crystallized samples will also be evaluated e.g. using """"""""texture patterns"""""""" from small molecules to provide a better 3D sampling of the reciprocal lattice to include regions not usually accessed by the limited tilting limits of the electron microscope goniometer stage. Fiber diffraction data from drawn polymer samples will also be evaluated. The goals of this project, therefore, will be to optimize phase prediction for limited data sets and to complete partial phase determinations made with the most reliable invariant relationships. Development of constrained least-squares refinement procedures will also be necessary when series termination errors from a sparse data set precludes accurate definition of atomic positions.
| Xu, Hongliang; Hauptman, Herbert A (2004) Statistical approach to the phase problem. Acta Crystallogr A 60:153-7 |
| Hauptman, Herbert A; Langs, David A (2003) The phase problem in neutron crystallography. Acta Crystallogr A 59:250-4 |
| Shen, Qun; Wang, Jun (2003) Recursive direct phasing with reference-beam diffraction. Acta Crystallogr D Biol Crystallogr 59:809-14 |
| Xu, Hongliang; Hauptman, Herbert A (2003) On integrating the techniques of direct methods and SIRAS: the probabilistic theory of doublets and its applications. Acta Crystallogr A 59:60-5 |
| Weeks, Charles M; Adams, Paul D; Berendzen, Joel et al. (2003) Automatic solution of heavy-atom substructures. Methods Enzymol 374:37-83 |
| Shen, Qun; Wang, Jun; Ealick, Steven E (2003) Anomalous difference signal in protein crystals. Acta Crystallogr A 59:371-3 |
| Shen, Qun (2003) Improving triplet-phase accuracy by symmetry observations in reference-beam diffraction measurements. Acta Crystallogr A 59:335-40 |
| Lemke, Christopher T; Smith, G David; Howell, P Lynne (2002) S-SAD, Se-SAD and S/Se-SIRAS using Cu Kalpha radiation: why wait for synchrotron time? Acta Crystallogr D Biol Crystallogr 58:2096-101 |
| Xu, Hongliang; Hauptman, Herbert A; Weeks, Charles M (2002) Sine-enhanced Shake-and-Bake: the theoretical basis and applications to Se-atom substructures. Acta Crystallogr D Biol Crystallogr 58:90-6 |
| Hauptman, Herbert A; Guo, D Y; Xu, Hongliang et al. (2002) Algebraic direct methods for few-atoms structure models. Acta Crystallogr A 58:361-9 |
Showing the most recent 10 out of 72 publications
