W0088

Developments in the ab initio Phasing of Icosahedral Viruses. Stephen T. Miller*, James M. Hogle*^[daggerdbl], and David J. Filman^[daggerdbl]. *Committee on Higher Degrees in Biophysics, Harvard University, Cambridge MA, ^[daggerdbl]Dept. of Biological Chemistry & Molecular Pharmacology, Harvard Medical School, Boston MA

A genetic algorithm method for the de novo phasing of low resolution x-ray diffraction data from crystals of icosahedral viruses has been implemented and shown, in the test case of synthetic data from poliovirus empty capsids, to give sufficiently correct phases for successful phase extension to high resolution through conventional real space averaging. Without advance knowledge of the shape of the virus and only approximate knowledge of its size, the virus can be modeled as the symmetry-expansion of a short list of nearly tetrahedrally arranged lattice points which coarsely, but uniformly, sample the icosahedrally unique volume of the virus. This parameterization permits a simple matrix formulation of the model evaluation, yielding a highly efficient survey of the space of possible models. A genetic algorithm is used to quickly search this parameter space and generate a population of low resolution trial models. Later steps in the procedure, applied to the most promising candidates, eliminate the coarseness of the lattice of the initial parameterization. Finally, real space averaging is used to improve the low resolution model and the for phase extension [1].

After successfully assigning phases to 24Å synthetic data from poliovirus empty capsids and phase extending to 12Å data, work has focused on applying the procedure to real data from native poliovirus. Initial attempts at phase extension, using data from 80 to 24Å were unsuccessful, possibly due to ripples introduced in the transform by the truncation of the data at low resolution. Recently, a much more complete and accurate data set has been collected from a single, frozen crystal of poliovirus. All low resolution reflections were collected (including the 110 reflection, the lowest present in space group P2₁2₁2) and high resolution data has been collected to at least 2.4Å. It is hoped that the elimination of the low resolution truncation in the transform will make the ab initio phasing procedure more robust. Attempts to apply the genetic algorithm to this new data set are currently underway. Efforts to use a cryoEM reconstruction as a starting point for phase extension are also in progress, in collaboration with D. Belnap, B. Trus and A. Steven.

[1] S.T. Miller, J.M. Hogle, D.J. Filman (1996) Acta Cryst. D52, 235-251