The dynamic nature of computer models is discussed. In addition to varying the three dimensional position of atoms and the amount of detail presented in a given model, the computer is also able to change the position of the camera relative to the atoms of a protein model.Thus, thecameracould be positioned outsideof themodel (theusualcase) or it could be moved inside. From inside the protein model, internal atoms are immediately before the camera and interactions can be studied at close range. Atoms behind the camera are out of sight but could be brought into view by rotating the structure. The computer program discussed in this report constructs the peptide backbone of a protein, amino acids that contain ring systems, and the heme prosthetic group (myoglobin). In addition, the program varies the position of the camera and the positions of the atoms to provide unique views of protein models. Methods also are described for preparing animated and stereoscopic three dimensional presentations from the CRT models.
Molecular models of proteins, based on X-ray data (1,2) or computer simulation, provide considerable insight into the factors involved in generating specific secondary, tertiary and quaternary structures. The construction of conventional framework molecular models for structures comprising thousands of atoms has a number of limitations, such as long construction times, accuracy of bond lengths and bond angles between atoms, accuracy in positioning each atom in three dimensional space. and storage facilities for large numbers of models. Furthermore, the resulting models are static since they are attached in numerous places to supporting structures. In contrast, the computer generated cathode ray tube (CRT) model (3) provides a favorable alternative since it is generated in seconds, atoms are placed precisely in three dimensional space, bond lengths and bond angles are accurate, and large * This research was supported by the Public Health Service (GRSG) and by the Atomic Energy Commission (B.N.L.).