A package of programs has been developed for efficient restrained least-squares refinement of macromolecular crystal structures. The package has been designed to be as flexible and general purpose as possible. The process of refinement is divided into basic units and an independent computer program * Present address: 79521 Highway 99N, Cottage Grove, Oregon 97424, USA.handles each task. Each functional unit communicates with other programs in the package by way of files of well defined format. To modify or replace any program, the user need only understand the function of that particular element. Stereochemical restraints are defined in a general way that can be applied to proteins, nucleic acids, prosthetic groups, solvent atoms and so on. Guide values for bond lengths and bond angles are specified in a straightforward direct manner. constant or constrained to behave as a rigid body during refinement. In order to make the package as efficient as possible, the fast Fourier transform algorithm is used for all the crystallographic transformations. To highlight potential errors in the refined structure the user can list those atoms that have the worst bond lengths and angles, or have the largest positional, temperature-factor or occupancy gradients. It is also possible to check that protein and solvent atoms do not sterically clash with symmetryrelated neighbors. Applications of the program package to a bacteriochlorophyll-containing protein, thermolysin-inhibitor complexes and mutants of bacteriophage T4 lysozyme are described.

No abstract

No abstract

To understand better how the packing of side chains within the core influences protein structure and stability, the crystal structures were determined for eight variants of T4 lysozyme, each of which contains three to five substitutions at adjacent interior sites. Concerted main-chain and side-chain displacements, with movements of helical segments as large as 0.8 angstrom, were observed. In contrast, the angular conformations of the mutated side chains tended to remain unchanged, with torsion angles within 20° of those in the wild-type structure. These observations suggest that not only the rotation of side chains but also movements of the main chain must be considered in the evaluation of which amino acid sequences are compatible with a given protein fold.