Mean-field minimization methods for biological macromolecules

Current Opinion in Structural Biology

Volumn 6, Issue 2, 1996, Pages 222-226

  Koehl, Patrice ^a   Delarue, Marc ^b  

^a UNIVERSITÉ DE STRASBOURG   (France)

^b INSTITUT PASTEUR   (France)

Author keywords

[No Author keywords available]

Indexed keywords

AMINO ACID SEQUENCE; MOLECULAR MODEL; PRIORITY JOURNAL; PROTEIN CONFORMATION; PROTEIN FOLDING; SHORT SURVEY;

EID: 0029869187     PISSN: 0959440X     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0959-440X(96)80078-9     Document Type: Article

References (36)

1. Gibson KD, Scheraga HA: The multiple minima problem in protein folding. In Structure and Expression: from Proteins to Ribosomes. Edited by Sarma RH, Sarma MH. Guilderland, NY: Adenine Press; 1988:67-94.
2. Dili KA, Bromberg S, Yue KZ, Fiebig KM, Yee DP, Thomas PD, Chan HS: Principles of protein folding - a perspective from simple exact models. Protein Sci 1995, 4:561-602. A thorough review on exact hydrophobic-polar (HP) lattice models of proteins, and their use in studying the thermodynamics and kinetics of protein folding processes.
3. Elber R, Karplus M: Enhanced sampling in molecular dynamics: use of the time-dependent Hartree approximation for a simulation of carbon monoxide diffusion through myoglobin. J Am Chem Soc 1990, 112:9161-9175.
4. Finkelstein AV, Reva BA: A search for the most stable folds of protein chains. Nature 1991, 351:497-499.
5. Roitberg A, Elber R: Modelling side-chains in peptides and proteins: application of the locally enhanced sampling and the simulated annealing method to find minimum energy conformations. J Chem Phys 1991, 95:9277-9287.
6. Koehl P, Delarue M: Application of a self consistent mean field theory to predict protein side-chain conformations and estimate their conformational entropy. J Mol Biol 1994, 239:249-275. A method for modelling protein side chains on a given framework is described. The method is based on a rotamer library and iteratively refines a conformational matrix of the side chains of a protein, using a self-consistent mean-field approach. The energy function only includes 12-6 Lennard Jones potentials. The rotamers with the highest probability in the optimized conformation matrix are used for final prediction. The conformational matrix also provides estimates of the configurational entropy of the side chains in folded proteins (see also [7••,9••])
7. Lee C: Predicting protein mutant energetics by self consistent ensemble optimisation. J Mol Biol 1994, 236:918-939. A self-consistent ensemble optimization is applied to predict side-chain conformations in the core of a protein and the effect of mutations on protein stability. It differs from the method presented in [6••] in that it does not rely on a rotamer library. Its application, however, is limited to small proteins, or to a limited number of side chains in a larger system (see also [ 9••]).
8. Zheng Q, Kyle DJ: Multiple copy sampling: rigid versus flexible protein. Proteins 1994, 19:324-329. The effects of protein flexibility on multiple copy sampling are systematically evaluated by studying a single side chain placement. The technique is shown to be significantly more efficient when a flexible, but harmonically constrained, protein is used instead of a rigid protein.
9. Vásquez M: An evaluation of discrete and continuum search techniques for conformational analysis of side chains in proteins. Biopolymers 1995, 36:53-70. Methodologies for calculation of side-chain conformations in proteins are evaluated, including heat bath algorithm and MFT techniques. It is shown that using non-ideal rotamers, as well as modification of the Lennard Jones energy function to a 9-6 potential instead of a 12-6 potential, leads to more accurate predictions.
10. Olszewski KA, Piela L, Scheraga HA: Mean field theory as a tool for intramolecular conformational optimisation. 1. Tests on terminally-blocked alanine and met-enkephalin. J Phys Chem 1992, 96:4672-4676.
11. Olszewski KA, Piela L, Scheraga HA: Mean field theory as a tool for intramolecular conformational optimisation. 2. Tests on the homopolypeptides decaglycine and icosalanine. J Phys Chem 1993, 97:260-266.
12. Olszewski KA, Piela L, Scheraga HA: Mean field theory as a tool for intramolecular conformational optimisation. 3. Test on melittin. J Phys Chem 1993, 97:267-270.
13. Rabow AA, Scheraga HA: Lattice neural network minimization. Application of neural network optimization for locating the global-minimum conformations of proteins. J Mol Biol 1993, 232:1157-1168.
14. Finkelstein AV, Reva BA: Search for the stable state of a short chain in a molecular field. Protein Eng 1992, 5:617-624.
15. Zheng Q, Rosenfeld R, DeLisi C, Kyle JD: Multiple copy sampling in protein loop modeling: computational efficiency and sensitivity to dihedral angle perturbations. Protein Sci 1994, 3:493-506. The bond-scaling relaxation technique proposed earlier by some of the same authors is combined with multiple copy sampling to enable them to model loop conformations in proteins, leading to an increased efficiency of up to a factor of five. The variability in the converged loop conformations is used to estimate the accuracy of the models.
16. Koehl P, Delarue M: A self consistent mean field approach to simultaneous gap closure and side-chain positioning in homology modelling. Nature Struct Biol 1995, 2:163-170. The SCMF approach presented in [6••] is generalized to apply to the problem of the simultaneous modelling of loops and the positioning of side chains in proteins. It relies on a database search to generate possible fragments for modelling gaps, and on a rotamer library to define side-chain conformations. Each loop fragment and each side-chain rotamer is characterized with a probability, which is refined using MFT. We show that, in the presence of multiple loops, simultaneous modelling yields more accurate predictions (see [14]).
17. Reva BA, Finkelstein AV: A new approach to the design of a sequence with the highest affinity for a molecular surface. Protein Eng 1992, 5:625-628.
18. Kono H, Doi J: Energy minimization method using automata network for sequence and side-chain conformation prediction from given backbone geometry. Proteins 1994, 19:244-255. This paper presents the use of a method of energy minimization, using an automata network, to predict a set of amino acid sequences and their side-chain conformations compatible with a given backbone geometry. It is essentially similar to the lattice neural network minimization procedure of Rabow and Scheraga [13] and bears some resemblance to the MFT techniques. Applications are restricted to the interior core region, with limited choice for amino acid types (six hydrophobic residues). The optimized sequences were found to be as well packed as the native sequence.
19. Landau LD, Lifshitz EM: Quantum Mechanics. New York: Pergamon Press; 1958.
20. Czerminski R, Elber R: Computational studies of ligand diffusion in globins: 1. Leghemoglobin. Proteins 1991, 10:70-80.
21. Straub JE, Karplus M: Energy equipartitioning in the classical time-dependent Hartree approximation. J Chem Phys 1991, 94:6737-6739.
22. Rosenfeld R, Zheng Q, Vajda S, DeLisi C: Computing the structure of bound peptides: application to antigen recognition by class I MHCs. J Mol Biol 1993, 234:515-520.
23. Zheng Q, Rosenfeld R, Kyle DJ: Theoretical analysis of the multi-copy sampling method in molecular modeling. J Chem Phys 1993, 99:8892-8896.
24. Kubo R: Statistical Physics. Amsterdam: North Holland Publishing Co; 1965.
25. Van den Bout DE, Miller TKI: Improving the performance of the Hopfield-Tank neural network through normalization and annealing. Biol Cybern 1989, 62:129-139.
26. Peterson C, Söderberg B: A new method for mapping optimization problems onto neural networks. Internat J Neural Systems 1989, 1:3-22.
27. Hopfield JJ, Tank DW: "Neural" computation of decisions in optimization problems. Biol Cybern 1985, 52:141-152.
28. Tuffery P, Etchebest C, Hazout S, Lavery R: A new approach to the rapid determination of protein side-chain conformations. J Biomol Struct Dyn 1991, 8:1267-1289.
29. Ponder JW, Richards FM: Tertiary templates for proteins. Use of packing criteria in the enumeration of allowed sequences for different structural classes. J Mol Biol 1987, 193:775-791.
30. Jones DT: De novo protein design using pairwise potentials and a genetic algorithm. Protein Sci 1994, 3:567-574. This paper describes a method that is based on a genetic algorithm for designing protein sequences which fit a given fold. The search is based on a potential energy function which contains statistically derived pairwise contact potentials (i.e. potentials of mean force) and solvation energies. Meaningful sequences are only obtained if the amino acid composition is restrained.
31. Sasai M: Conformation, energy, and folding ability of selected amino acid sequences. Proc Natl Acad Sci USA 1995, 92:8438-8442. This paper describes a simulated annealing procedure that designs protein sequences compatible with a given fold. It shows that a naive energy minimization leads to unstable and non-folding sequences. Better sequences are obtained when the energy landscape over a wide region of the conformational space is imposed. The hydrophobicity pattern and the glycine locations are conserved among the foldable sequences.
32. Hellinga HW, Richards FM: Optimal sequence selection in proteins of known structure by simulated evolution. Proc Natl Acad Sci USA 1994, 91:5803-5807. This paper describes a computational method that designs sequences that fit in a given protein core, using a simulated-annealing approach in which random walks over sequence and configurational space are performed. The potential energy function driving the search includes terms for van der Waals non-bonded interactions, solvation energies and entropy. Applications to the core of the phage λ c1 repressor are presented.
33. Godzik A: In search of the ideal protein sequence. Protein Eng 1995, 8:409-416. This paper describes a Monte Carlo procedure that designs protein sequences that fit a given fold. The native amino acid composition is imposed. To derive foldable sequences, the procedure optimizes the 'energy gap', that is, the difference in energy for a given sequence between its native and the best non-native conformations.
34. Wodak SJ, Rooman MJ: Generating and testing protein folds. Curr Opin Struct Biol 1993, 3:247-259.
35. Wang Y, Zhang H, Li W, Scott RA: Discriminating compact nonnative structures from the native structure of globular proteins. Proc Natl Acad Sci USA 1995, 92:709-713.
36. Burling FT, Brünger AT: Thermal motion and conformational disorder in protein crystal structures: comparison of multiconformer and time-averaging models. Isr J Chem 1994, 34:165-175. To account for thermal motion during the refinement of a protein crystal structure, three methods were tested: conventional refinement, multicopy sampling and time-averaging refinement using molecular dynamics. In the case of penicillopepsin at 1.8 Å, the best results were obtained with simultaneous refinement of between four and eight conformers.