Computational folding of proteins moved in this year from an abstract geometry-free model developed during the previous year to a 3- dimensionally embedded constraint evaluation model. The geometry-free model was implemented as a series of topological connections between charged atoms representing the hydrophilic aspects of peptide biochemistry and another series of connections between groups of carbon atoms representing the hydrophobic aspects. The rule-based manipulation of topological connections is, computationally, relatively inexpensive. In order to maintain a model that is physically reasonable, the sequential synthesis of the protein from N to C terminus is modeled. In ribosomal synthesis, it is only when the peptide emerges from the ribosomal that it begins to adopt a folded conformation. The linearly increasing peptide length keeps the DGEOM computation time to a minimum. We have found that the pattern of hydrophilic and hydrophobic constraints develops with the lengthening of the peptide in such a way that the peptide adopts conformations that are very close to the crystal or NMR structure observed after ribosomal emission. As the sequence for a helical portion of a peptide is emitted, it folds into a helix. As soon as the sequence for an antiparallel beta sheet is emitted, it too folds into the correct secondary structure. The most striking result from this year's simulations is the discovery that the strand-helix-strand peptide sequence forms a tertiary structure with the correct macroscopic handedness when it is emitted. Ten classes of protein architecture are being studied in parallel, in an attempt to make the computational simulation of protein folding as general as possible. We have observed that, in all structural classes, the correct local secondary structure is formed, and often the correct macroscopic handedness is also formed. So far, the simulation program has not been able to reproduce the tightly packed atomic structure characteristic of crystal NMR structures. Rules and parameters are continually added and modified, in attempts to produce better packing.

Agency
National Institute of Health (NIH)
Institute
Center for Information Technology (CIT)
Type
Intramural Research (Z01)
Project #
1Z01CT000236-04
Application #
3752815
Study Section
Project Start
Project End
Budget Start
Budget End
Support Year
4
Fiscal Year
1994
Total Cost
Indirect Cost
Name
Center for Information Technology
Department
Type
DUNS #
City
State
Country
United States
Zip Code