No abstract

Structural genomics seeks to expand rapidly the number of protein structures in order to extract the maximum amount of information from genomic sequence databases. The advent of several large-scale projects worldwide leads to many new challenges in the ®eld of crystallographic macromolecular structure determination. A novel software package called PHENIX (Python-based Hierarchical ENvironment for Integrated Xtallography) is therefore being developed. This new software will provide the necessary algorithms to proceed from reduced intensity data to a re®ned molecular model and to facilitate structure solution for both the novice and expert crystallographer.

Obtaining an electron-density map from X-ray diffraction data can be dif®cult and time-consuming even after the data have been collected, largely because MIR and MAD structure determinations currently require many subjective evaluations of the qualities of trial heavy-atom partial structures before a correct heavy-atom solution is obtained. A set of criteria for evaluating the quality of heavy-atom partial solutions in macromolecular crystallography have been developed. These have allowed the conversion of the crystal structure-solution process into an optimization problem and have allowed its automation. The SOLVE software has been used to solve MAD data sets with as many as 52 selenium sites in the asymmetric unit. The automated structure-solution process developed is a major step towards the fully automated structure-determination, model-building and re®nement procedure which is needed for genomic scale structure determinations.

A likelihood-based approach to density modi®cation is developed that can be applied to a wide variety of cases where some information about the electron density at various points in the unit cell is available. The key to the approach consists of developing likelihood functions that represent the probability that a particular value of electron density is consistent with prior expectations for the electron density at that point in the unit cell. These likelihood functions are then combined with likelihood functions based on experimental observations and with others containing any prior knowledge about structure factors to form a combined likelihood function for each structure factor. A simple and general approach to maximizing the combined likelihood function is developed. It is found that this likelihood-based approach yields greater phase improvement in model and real test cases than either conventional solvent¯attening and histogram matching or a recent reciprocal-space solvent-¯attening procedure [Terwilliger (1999), Acta Cryst. D55, 1863±1871].

Existing protein tagging and detection methods are powerful but have drawbacks. Split protein tags can perturb protein solubility or may not work in living cells. Green fluorescent protein (GFP) fusions can misfold or exhibit altered processing. Fluorogenic biarsenical FLaSH or ReASH substrates overcome many of these limitations but require a polycysteine tag motif, a reducing environment and cell transfection or permeabilization. An ideal protein tag would be genetically encoded, would work both in vivo and in vitro, would provide a sensitive analytical signal and would not require external chemical reagents or substrates. One way to accomplish this might be with a split GFP, but the GFP fragments reported thus far are large and fold poorly, require chemical ligation or fused interacting partners to force their association, or require coexpression or co-refolding to produce detectable folded and fluorescent GFP. We have engineered soluble, self-associating fragments of GFP that can be used to tag and detect either soluble or insoluble proteins in living cells or cell lysates. The split GFP system is simple and does not change fusion protein solubility.