A new software routine that automates the fitting of protein X-ray crystallographic electron-density maps.

The classical approach to building the amino-acid residues into the initial electron-density map requires days to weeks of a skilled investigator's time. Automating this procedure should not only save time, but has the potential to provide a more accurate starting model for input to refinement programs. The new software routine MAID builds the protein structure into the electron-density map in a series of sequential steps. The first step is the fitting of the secondary alpha-helix and beta-sheet structures. These 'fits' are then used to determine the local amino-acid sequence assignment. These assigned fits are then extended through the loop regions and fused with the neighboring sheet or helix. The program was tested on the unaveraged 2.5 A selenomethionine multiple-wavelength anomalous dispersion (SMAD) electron-density map that was originally used to solve the structure of the 291-residue protein human heart short-chain L-3-hydroxyacyl-CoA dehydrogenase (SHAD). Inputting just the map density and the amino-acid sequence, MAID fitted 80% of the residues with an r.m.s.d. error of 0.43 A for the main-chain atoms and 1.0 A for all atoms without any user intervention. When tested on a higher quality 1.9 A SMAD map, MAID correctly fitted 100% (418) of the residues. A major advantage of the MAID fitting procedure is that it maintains ideal bond lengths and angles and constrains phi/psi angles to the appropriate Ramachandran regions. Recycling the output of this new routine through a partial structure-refinement program may have the potential to completely automate the fitting of electron-density maps.

[1]  J Greer.  Three-dimensional pattern recognition: an approach to automated interpretation of electron density maps of proteins. , 1974, Journal of molecular biology.

[2]  A. Abbott.  Structures by numbers , 2000, Nature.

[3]  S. Doublié.  [29] Preparation of selenomethionyl proteins for phase determination. , 1997, Methods in enzymology.

[4]  L. Banaszak,et al.  Biochemical characterization and crystal structure determination of human heart short chain L-3-hydroxyacyl-CoA dehydrogenase provide insights into catalytic mechanism. , 1999, Biochemistry.

[5]  B. L. Sibanda,et al.  Conformation of beta-hairpins in protein structures. A systematic classification with applications to modelling by homology, electron density fitting and protein engineering. , 1989, Journal of molecular biology.

[6]  J. Zou,et al.  Improved methods for building protein models in electron density maps and the location of errors in these models. , 1991, Acta crystallographica. Section A, Foundations of crystallography.

[7]  J. Thornton,et al.  PROCHECK: a program to check the stereochemical quality of protein structures , 1993 .

[8]  E. Haug,et al.  A recursive formulation constrained mechanical system dynamics. I: Open loop systems , 1987 .

[9]  Thomas C Terwilliger,et al.  [30] Multiwavelength anomalous diffraction phasing of macromolecular structures: Analysis of MAD data as single isomorphous replacement with anomalous scattering data using the MADMRG program. , 1997, Methods in enzymology.

[10]  D. Levitt,et al.  A new routine for thinning, editing and fitting MIR maps using real-space molecular dynamics , 1993 .

[11]  W. Hendrickson.  Determination of macromolecular structures from anomalous diffraction of synchrotron radiation. , 1991, Science.

[12]  P. Smaglik.  Genomics initiative to decipher 10,000 protein structures , 2000, Nature.

[13]  Roland L. Dunbrack,et al.  Backbone-dependent rotamer library for proteins. Application to side-chain prediction. , 1993, Journal of molecular biology.

[14]  K D Cowtan,et al.  Density modification for macromolecular phase improvement. , 1999, Progress in biophysics and molecular biology.