pubmed.ncbi.nlm.nih.gov

Improvement of molecular-replacement models with Sculptor - PubMed

Improvement of molecular-replacement models with Sculptor

Gábor Bunkóczi et al. Acta Crystallogr D Biol Crystallogr. 2011 Apr.

Abstract

In molecular replacement, the quality of models can be improved by transferring information contained in sequence alignment to the template structure. A family of algorithms has been developed that make use of the sequence-similarity score calculated from residue-substitution scores smoothed over nearby residues to delete or downweight parts of the model that are unreliable. These algorithms have been implemented in the program Sculptor, together with well established methods that are in common use for model improvement. An analysis of the new algorithms has been performed by studying the effect of algorithm parameters on the quality of models. Benchmarking against existing techniques shows that models from Sculptor compare favourably, especially if the alignment is unreliable. Carrying out multiple trials using alternative models created from the same structure but using different algorithm parameters can significantly improve the success rate.

PubMed Disclaimer

Figures

Figure 1
Figure 1

Graphical representation of the parameters used in sequence-similarity calculations.

Figure 2
Figure 2

Sequence-identity distribution of models used in the benchmark suite. Solution statistics are indicated for ClustalW alignments. Reference, solved by the reference protocol (protocol 1); cumulative, solved by any protocols; unsolved, no solutions found.

Figure 3
Figure 3

Sequence-similarity scores calculated with various settings for an insertion in the model (PDB entry

2b9l

; Piao et al., 2005 ▶). The target structure (PDB entry

1hj9

; Leiros et al., 2001 ▶) is shown in grey. Sequence similarity was calculated using (a) a high-precision structural alignment (target, HCY––––––––––KSGIQVR; model, HCVNSYQSNLDA­I––––KIR) using the binary scoring matrix and a null averaging window, or an FFAS alignment (target, HCYKS––––––GIQVR; model, HCV­NSYQSNLDAIKIR) and (b) the binary matrix and a null averaging window, (c) BLOSUM62 and a null averaging window or (d) BLOSUM62 and an averaging window of five. Blue indicates high sequence similarity, while red indicates areas in which sequence similarity is low. The figures were generated using MOLSCRIPT (Kraulis, 1991 ▶) and were rendered with RASTER3D (Merritt & Bacon, 1997 ▶).

Figure 4
Figure 4

Sequence-similarity scores calculated with various settings for a deletion in the model (PDB entry

1hj9

). The target structure (PDB entry

2b9l

) is shown in grey. Alignments are identical to those used for Fig. 3 ▶, but the roles of target and model are reversed. Sequence similarity was calculated using (a) a high-precision structural alignment using the binary scoring matrix and a null averaging window, or an FFAS alignment and (b) the binary matrix with a null averaging window, (c) BLOSUM62 and a null averaging window or (d) BLOSUM62 and an averaging window of five. Blue indicates high sequence similarity, while red indicates areas in which sequence similarity is low. The figures were generated using MOLSCRIPT (Kraulis, 1991 ▶) and rendered with RASTER3D (Merritt & Bacon, 1997 ▶).

Similar articles

Cited by

References

    1. Adams, P. D. et al. (2010). Acta Cryst. D66, 213–221. - PubMed
    1. Altschul, S. F., Gish, W., Miller, W., Myers, E. W. & Lipman, D. J. (1990). J. Mol. Biol. 215, 403–410. - PubMed
    1. Berman, H. M. et al. (2002). Acta Cryst. D58, 899–907. - PubMed
    1. Dayhoff, M. O., Schwartz, R. M. & Orcutt, B. C. (1978). Atlas of Protein Sequence and Structure, Vol. 5, edited by M. O. Dayhoff, pp. 345–352. Washington DC: National Biomedical Research Foundation.
    1. Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. (2010). Acta Cryst. D66, 486–501. - PMC - PubMed

Publication types

MeSH terms