Biomedical subjects
Publications and source records attributed to Giampiero Polidori.
Two algorithms are described for limiting data resolution and for predicting the most informative wavelength combinations in MAD techniques. Both have been successfully tested using experimental data from a large set of test structures.
New algorithms for solving ab initio protein crystal structures have been identified and implemented in a modified version of the program SIR2002. They succeed in solving numerous protein structures diffracting at atomic resolution; the solution was also attained when data were cut at 1.4 A resolution. The direct-space refinement procedure of SIR2003-N takes advantage of using the envelope of the protein, calculated during the phasing process from the current phases. The electron-density map is modified by assuming different weights for pixels within the envelope or out of it, so tentatively depleting the intensities of the false peaks. The map is then inverted and the resulting phase sets may improve their values. The new phasing strategy is also based on an optimal use of some figures of merit, one of which may be successfully applied in the early stages of the phasing process: only the most promising trials are submitted to the complete phasing procedure, so saving computing time. SIR2003-N has been successfully applied also in solving some protein structures diffracting at 1.4-1.5 A resolution.
All the techniques today available for the ab initio crystal structure solution of proteins require that the atomicity condition is satisfied. Accordingly, diffraction data at resolution equal or better than 1.2 A are necessary. This condition reduces the role of the ab initio techniques in macromolecular crystallography. The computer program SIR2002 has been modified in such a way that it may succeed also with 1.4 A resolution diffraction data. The modifications concern all the modules of the program: the modified program also benefits by the efficiency of a figure of merit.
The method of joint probability distribution functions is applied in order to estimate the structure-factor moduli of the anomalous scatterer substructure both in the SAD (single-wavelength anomalous dispersion) and in the MAD (multi-wavelength anomalous dispersion) cases. The experimental data |F(1)(+)|, |F(1)(-)|, ..., |F(n)(+)|, |F(n)(-)| measured at n wavelengths are used simultaneously to estimate the value of |F(oa)| arising from the normal scattering of the anomalous scatterers. A practical procedure is described that, when applied to the experimental diffraction data of several proteins, shows robustness and efficiency.
The method of the joint probability distribution function is applied in order to estimate the structure-factor moduli of the anomalous scatterer substructure. The two-wavelength case is examined: the prior knowledge of the moduli [F(1)(+)], [F(1)(-)], [F(2)(+)], [F(2)(-)] is used to predict the value of [F(oa)] arising from the normal scattering of the anomalous scatterers. The conclusive formula is applied to ideal and to real cases: evidence of the usefulness of the approach is obtained.