Protein structure: What is it possible to predict now?

Current Opinion in Structural Biology

Volumn 7, Issue 1, 1997, Pages 60-71

  Finkelstein, Alexei V ^a  

^a INSTITUTE OF PROTEIN RESEARCH   (Russian Federation)

Author keywords

[No Author keywords available]

Indexed keywords

PROTEIN;

ANALYTICAL ERROR; MOLECULAR DYNAMICS; PRIORITY JOURNAL; PROTEIN CONFORMATION; PROTEIN FOLDING; PROTEIN INTERACTION; PROTEIN STABILITY; PROTEIN STRUCTURE; REVIEW;

EID: 0031052179     PISSN: 0959440X     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0959-440X(97)80008-5     Document Type: Article

References (113)

1. Schulz GE, Barry CD, Friedman J, Chou PY, Fasman GD, Finkelstein AV, Lim VI, Ptitsyn OB, Kabat EA, Wu TT, et al. Comparison of predicted and experimentally determined secondary structure of adenilate kinase. Nature. 250:1974;140-142.
2. Ptitsyn OB, Rashin AA. A model of myoglobin self-organisation. Biophys Chem. 3:1975;1-20.
3. Cohen FE, Sternberg MJE, Taylor WR. Analysis and prediction of protein β-sheet structures by a combinatorial approach. Nature. 285:1980;378-382.
4. Ptitsyn OB, Finkelstein AV. Self-organisation of proteins and the problem of their three-dimensional structure prediction. Jaenicke R. Protein Folding. 1980;101-115 Elsevier, Amsterdam.
5. Mornon JP, Fridlansky E, Bally R, Milgrom E. X-ray crystallographic analysis of a progesterone binding protein. The C222, crystal form of oxidised uteroglobin at 2.2 Å resolution. J Mol Biol. 137:1980;415-429.
6. Bennett MJ, Schlunegger MP, Eisenberg D. 3D domain swapping: a mechanism for oligomer assembly. Protein Sci. 4:1995;2455-2468.
7. Benner SA, Gerloff D. Patterns of divergence in homologues proteins as indicators of secondary and tertiary structure: the catalytic domain of protein kinases. Adv Enzyme Reg. 31:1990;121-181.
8. Bowie JU, Luthy R, Eisenberg D. A method to identify protein sequences that fold into a known three-dimensional structure. Science. 253:1991;164-170.
9. Finkelstein AV, Reva BA. A search for the most stable folds of protein chains. Nature. 351:1991;497-499.
10. Thornton JM, Flores TP, Jones DT, Swindells MB. Prediction of progress at last. Nature. 354:1991;105-106.
11. of special interest Moult J, Judson R, Fidelis K, Pedersen JT. Large scale experiment to assess protein structure prediction methods. Proteins. 23:1995;ii-iv This is a short account of the Meeting on the Critical Assessment of Protein Structure Prediction Methods ([CASP]-1; 1994 December 4-8, Asilomar, USA) and a preface to the proceedings of the meeting that are published in the same special issue of Proteins.
12. Protein Structure Prediction Center on World Wide Web URL: http://Prediction center.IInl.gov.
13. of special interest Lemer CMR, Rooman MJ, Wodak SJ. Protein structure prediction by threading methods: evaluation of current techniques. of outstanding interest Proteins. 23:1995;337-355 This paper contains a comprehensive analysis of the results of the Asilomar contest [11] in 'threading'. Threading is a technique that aligns a protein sequence with a set of known 3D structures to select the correct fold of this sequence from a set of alternatives. It is shown that when the native fold of the examined chain is close to some fold from the set, the threading methods can more or less often (but not always) recognize the correct fold as one of the most probable ones for this sequence. The sequence-structure alignments in all the correctly recognized folds, however, are suprisingly poor, in other words, threading cannot currently derive a detailed 3D model from the amino-acid sequence. It is suggested that the fold recognition is actually based upon recognition of only rough structural features: secondary structures and hydrophobic interactions in the protein core. The review also contains a brief but very informative classification of threading methods.
14. Mosiman S, Melishko R, James MGN. A critical assessment of comparative molecular modelling of tertiary structure of proteins. of special interest Proteins. 23:1995;301-317 The assessment shows that comparative molecular modelling is highly successful when sequence identify exceeds 70%, and is poor when the identity is below 30%. Where the sequence alignment is in error (e.g. in regions of poor local homology) the comparative model is guaranteed to be wrong. The most serious errors occur in the regions of insertions and deletions.
15. Defay T, Cohen FE. Evaluation of current techniques for ab initio protein structure prediction. of outstanding interest Proteins. 23:1995;431-445 This paper reviews the results of an ab initio protein-structure prediction contest from the level of secondary structure to the level of 3D folds. The results show that ab initio prediction of novel folds and, in general, an accurate tertiary-structure prediction is not yet possible. Protein fold and motif recognition are possible, however, when the motif is sufficiently similar to another, already known protein structure. The predictions are facilitated by aligned families of homologues sequences. The paper includes a very useful summary of the ab initio prediction techniques used for the Asilomar contest [11].
16. Hubbard T, Tramontano A. Update on protein structure prediction: results of the 1995 IRBM workshop. of special interest Fold Des. 1:1996;R55-R63 This is an account of a practical workshop on protein-structure prediction with a very useful list of software used in the course of predictions.
17. Levinthal C. Are there pathways for protein folding? J Chim Phys Phys Chim Biol. 65:1968;44-45.
18. Anfinsen CB. Principles that govern the folding of protein chains. Science. 181:1973;130-223.
19. Finkelstein AV, Galzitskaya OV, Badretdinov AY. A folding pathway solving Levinthal's paradox [abstract]. 12th International Biophysics Congress. 1996 August 11-16;. Amsterdam.
20. Privalov PL. Stability of proteins. Small globular proteins. Adv Prot Chem. 33:1979;167-241.
21. Privalov PL. Stability of proteins. Proteins which do not present a single cooperative unit. Adv Prot Chem. 35:1982;1-104.
22. Goldstein RA, Luthey-Schulten ZA, Wolynes PG. Optimal protein-folding codes from spin-glass theory. Proc Natl Acad Sci USA. 89:1992;4918-4922.
23. Finkelstein AV. Implementation of the random characteristics of protein sequences for their three-dimensional structures. Curr Opin Struct Biol. 4:1994;422-428.
24. Finkelstein AV, Gutin AM, Badretdinov AY. Boltzmann-like statistics of protein architectures. Origins and consequences. Biswas BB, Roy S. Subcellular Biochemistry. Proteins: Structure, Function and Protein Engineering. 24:1995;1-26 Plenum Press, New York.
25. Ngo JT, Marks J. Computational complexity of a problem of molecular structure prediction. Protein Eng. 5:1992;313-321.
26. Lathrop RH. The protein threading problem with sequence amino acid preferences is NP-complete. Protein Eng. 7:1994;1059-1068.
27. Brooks BR, Bruccoleri RE, Olafson BD, States DJ, Swaminathan S, Karplus M. CHARMM: a program for molecular energy minimisation and dynamics calculations. J Comp Chem. 4:1983;187-217.
28. Levitt M, Warshel. A computer simulation of protein folding. Nature. 253:1975;694-698.
29. Skolnick J, Kolinski A. Monte-Carlo lattice dynamics and the prediction of protein folds. of special interest Van Gunstren WF, Weinevandt PK, Willeinson AJ. Computer Simulations of Biomolecular Systems, Theoretical and Experimental. 1997;ESCOM Science Publications, Leiden, A comprehensive review of work on Monte-Carlo lattice simulations of protein folding.
30. Rost B, Sander C. Progress of 1D protein structure prediction at last. of special interest Proteins. 23:1995;295-300 Based on the Asilomar critical assessment [11], this paper shows that the accuracy of secondary-structure predictions has improved significantly (up to 72±9% for correctly predicting the α, β, or coil state of a residue) by using information contained in multiple sequence alignments as input to a neural network system [31].
31. Rost B. PHD: predicting one-dimensional protein structure by profile-based neural networks. Methods Enzymol. 266:1996;525-539.
32. Srinivasan R, Rose GD. LINUS - a hierarchical procedure to predict the fold of a protein. Proteins. 22:1995;81-99.
33. Benner SA, Gerloff D, Chelvanayagam G. The phosphogalactosidase and synaptotagmin prediction. of special interest Proteins. 23:1995;446-453 The paper analyzes the authors' blind predictions of protein structures. The predictions were carried out as follows: first, a secondary structure was predicted as a consensus structure from the multiple sequence alignments; and second, a tertiary structure was chosen by threading of the predicted secondary structure onto proteins of known 3D structure.
34. Bazan FJ. Helical fold prediction for the cyclin box. Proteins. 24:1996;1-17.
35. Gerloff DL, Cohen FE. Secondary structure prediction and unrefined tertiary structure prediction for the cyclin A, B and D. of special interest Proteins. 24:1996;18-34 Blind predictions of protein structures are presented. The 'Note added in proof' compares the authors' and Bazan's [34] predictions with the subsequently resolved cyclin crystal structure. Although the secondary structure predictions (both obtained from multiple sequence alignments) are extremely good (especially Bazan's one), the predictions of tertiary structures did not turn out to be successful.
36. Ptitsyn OB. Sequential mechanism of protein folding. Dokl Akad Nauk. 210:1973;1213-1215.
37. Sun S, Thomas PD, Dill KA. A simple protein folding algorithm using a binary code and secondary structure constraints. of special interest Protein Eng. 8:1995;769-778 The algorithm discussed predicts the tertiary structures of small proteins from their sequences, and their secondary structures using a small number of energy parameters.
38. Pedersen JT, Moult J. Ab initio structure prediction for small polypeptides and protein fragments using genetic algorithm. Proteins. 23:1995;454-460.
39. Bernstein FC, Koetzle TF, Meyer EF Jr, Brice MD, Kennard O, Shimanouchi T, Tasumi T. The protein databank: a computer-based archive file for macromolecular structures. J Mol Biol. 112:1977;535-542.
40. Novotny J, Rashin AA, Bruccoleri R. Criteria that discriminate between native proteins and incorrectly folded models. Proteins. 4:1988;14-30.
41. Hendlich M, Lackner P, Weitckus S, Flöckner H, Froschauer S, Gottsbacher K, Casari G, Sippl MJ. Identification of native protein folds against a large number of incorrect models. J Mol Biol. 216:1990;167-180.
42. s′ is the energy dispersion for subsequence s′ computed by a no-gap threading of s′ onto a control set of protein folds: the higher the z-score, the more probable is the alignment, and the more probable is that the native fold of the investigated sequence is similar to the given target fold.
43. Miyazava S, Jernigan RL. Residue - residue potentials with a favourable contact pair term and an unfavourable high packing density term, for simulation and threading. of special interest J Mol Biol. 256:1996;623-644 This work contains a refinement of the attractive interresidue contact energies and introduces an additional repulsive-packing energy term to prevent overlapping. The potentials are tested using no-gap threading experiments where 73 of 88 proteins demonstrate the lowest energy for the native folds.
44. Finkelstein AV, Reva BA. Computation of a chain fold in a globular protein by a self-consistent field method. Biofisika (USSR). 35:1990;401-406.
45. Jones DT, Taylor WR, Thornton JM. A new approach to protein fold recognition. Nature. 358:1992;86-89.
46. Sippl MJ, Weitckus S. Detection of native-like models for amino acid sequences of unknown three-dimensional structure in a data base of known protein conformations. Proteins. 13:1992;258-271.
47. Godzik A, Kolinski A, Skolnick J. Topology finger-print approach to the inverse protein folding problem. J Mol Biol. 227:1992;227-238.
48. Bryant SH, Laurence CE. An empirical energy function for threading protein sequence through the folding motif. Proteins. 16:1993;92-112.
49. Russel RB, Barton GJ. Structural features can be unconserved in proteins with similar folds. An analysis of side-chain to side-chain contacts, secondary structure and accessibility. J Mol Biol. 244:1994;332-350.
50. Finkelstein AV, Ptitsyn OB. Why do globular proteins fit the limited set of folding patterns? Progr Biophys Mol Biol. 50:1987;171-190.
51. Murzin AG, Finkelstein AV. General architecture of α-helical globules. J Mol Biol. 204:1988;749-770.
52. Murzin AG, Brenner SE, Hubbard TJP, Chothia C. SCOP: a structural classification of protein database for the investigation of sequences and structures. of special interest J Mol Biol. 247:1995;534-540 The paper presents a description of the database containing a comprehensive classification of protein folds, superfamilies, families, etc.
53. Madej T, Gibart JF, Bryant SH. Threading a database of protein cores. of outstanding interest Proteins. 23:1995;356-369 A critical analysis of the authors' blind protein-fold recognitions is presented which stresses that a successful fold recognition depends on the accurate definition of those parts of the fold that are, or are not, likely to be conserved in protein evolution.
54. Finkelstein AV, Reva BA. Search for the most stable folds of protein chains. I. Application of a self-consistent molecular field theory to a problem of protein three-dimensional structure prediction. of special interest Protein Eng. 9:1996;387-397 The paper presents a general approach to the prediction of 3D folds of protein chains from their amino-acid sequences. It is based on the use of the SCMF theory for long-range interactions, on the use of 1D statistical mechanics for short-range interactions, and on the fact that there are only a relatively small set of 'potentially stable' folding patterns. This makes it possible to examine the whole variety of potentially stable folds and to find the thermodynamically stable structure. The paper contains a detailed presentation of a SCMF theory as applied to protein chains and (as an Appendix) the computer algorithms of 1D statistical mechanics.
55. Reva BA, Finkelstein AV. Search for the most stable folds of protein chains. II. Computation of stable architectures of β-proteins using a self-consistent molecular field theory. of special interest Protein Eng. 9:1996;397-411 In this paper, a general SCMF approach is applied to single out the most stable fold for the chains of different β-sandwich proteins. The choice is made from the library of 96 different idealized β-sandwich-folding patterns. The native pattern is found out of 1-3, or sometimes out of 6, patterns having the lowest free energy for this chain (however, the results are much worse for the chains that have very long loops in their native folds). Computing a detailed chain fold within its native folding pattern, one obtains a rather close (although not perfect) similarity of calculated and observed structures.
56. Bateman A, Chothia C. Outline structures for the extracellular domains of the fibroblast growth factor receptors. of special interest Nat Struct Biol. 2:1995;1068-1074 The work presented is based on the 'key residues' that are identified in the known 3D structure as those primarily responsible for the structure of the central core of the protein through their packing, hydrogen-bonding or unusual conformations.
57. Wodak CJ, Rooman MJ. Generating and testing protein folds. Curr Opin Struct Biol. 3:1993;247-259.
58. Fisher D, Rice D, Bowie JU, Eisenberg D. Assigning amino acid sequences to 3-dimensional protein folds. of outstanding interest FASEB J. 10:1996;126-136 This is a review of the different approaches to fold recognition. The main steps are singled out: creation of a library of 'target' folds; representation of folds; alignment of the probe sequence to the target fold using the sequence-to-fold compatibility function; assessment of significance of compatibility. Various techniques used at each step are compared and the main technical problems to be solved are outlined.
59. of special interest Jones DT, Thornton J. Potential energy functions for threading. of outstanding interest Curr Opin Struct Biol. 6:1996;210-216 A very clearly written review of the different potential functions used for recognition of tertiary folds. The review concerns both the most simplified residue-energy functions and the sophisticated full-atom ones. The authors stress that although the pairwise interaction potentials mostly encode the residue-solvation preferences, the threading methods based on pairwise interactions work better than the profile-based (i.e. solvation) methods because accessibility is not well conserved in similar folds. Together with [87], the review underlines a parallel between experimental protein folding to a molten-globule state and fold prediction using simplified energy functions. The review ends with the phrase "In experimental folding, attaining correct sidechain packing appears to be rate-limiting step. Let's hope this is not the case for prediction". I am afraid that I believe this hope is weak.
60. Johnson MS, Overington J, Blundell TL. Alignment and search for the common protein folds using a data bank of structural templates. J Mol Biol. 231:1993;735-752.
61. Ouzounis C, Sander C, Scharf M, Schneider R. Prediction of protein structure by evaluation of sequence-structure fitness. Aligning sequences to contact profiles derived from three-dimensional structures. J Mol Biol. 232:1993;805-825.
62. Kocher JPA, Rooman MJ, Wodak SJ. Factors influencing the ability of knowledge-based potentials to identify native sequence-structure matches. J Mol Biol. 235:1994;1598-1613.
63. Matsuo Y, Nakamura H, Nishikava K. Detection of protein 3D - 1D compatibility characterised by the evaluation of side-chain packing and electrostatic interactions. of special interest J Biochem (Tokyo). 118:1995;137-148 The central point of this interesting but rather complicated paper is an attempt to take into account the sidechain packing in protein-fold recognition, that is, to take into account not only the interresidue distance but also the orientation of sidechains.
64. Matsuo Y, Nishikava K. Assessment of protein fold recognition method that takes into account four physicochemical properties: side-chain packing, solvation, hydrogen-bonding and local conformations. of special interest Proteins. 23:1995;370-375 See annotation [65].
65. Jones DT, Miller L, Thornton JM. Successful protein fold recognition by optimal sequence threading validated by rigorous blind test. of special interest Proteins. 23:1995;387-397 This paper, together with [64], analyzes the results of the authors' blind protein-structure recognitions.
66. Wilmanns M, Eisenberg D. Inverse protein folding by the residue pair preference method: estimating the correctness of alignments of structurally compatible sequences. of special interest Protein Eng. 8:1995;627-639 The method described in this paper combines environmental profiles with the pair-preference based profile; the latter is adjusted in the course of iterative computations. Testing shows that this method gives a 50% (on average) correct alignment when the rmsd between the native folds of target and probe chains is less than 2.
67. Jaroszewski L, Godzik A. Evaluation of threading alignments. of special interest Protein Eng. 1997; This paper describes a modification of the original 'fingerprint' approach to modelling [47]. The amino-acid environment of each point of the target fold is updated in the course of threading. The authors stress that the choice of gap insertion/deletion penalties is crucial for the quality of threading, and that this problem is yet unsolved. It is shown that sequence alignments are less accurate than threading alignments for distantly related proteins that have a sequence identity of 30% or less.
68. Pohl FM. Empirical protein energy maps. Nature. 234:1971;277-279.
69. Finkelstein AV, Badretdinov AY, Gutin AM. Why do protein architectures have a Boltzmann-like statistics? of special interest Proteins. 23:1995;142-150 A theoretical investigation of the typical statistics of the stable folds of random heteropolymers is performed to explain the basis of a Boltzmann-like form of the statistics of globular proteins. It is shown that the statistics resemble more the correlation functions observed in liquids or solids, in other words, they reflect the energy of interactions within a given element (which dominates when the internal interactions are strong), plus the interaction of this element with the environment. The 'temperature' in these statistics is the freezing temperature of the heteropolymers. The origin of these statistics is simply that the more stable is the element, the more sequences can provide stability to the folds with the element, and, as a result, the more often this element is observed in a random or quasirandom soup of sequences.
70. Thomas PD, Dill KA. Statistical potentials extracted from protein structures: how accurate they are? of special interest J Mol Biol. 257:1995;457-469 The results of this paper suggest that current statistical potentials may have a limited value: an exact lattice-model test has shown that these potentials often rank in order the true relative strengths of interresidue interactions correctly, but (at least for the investigated chains composed of two sorts of residues: in such chains the many-body effects are most visible [69,110]) they do not correctly reflect the true underlying energies.
71. Rooman M, Wodak S. Are database-derived potentials valid for the scoring both forward and inverted protein folding? of special interest Protein Eng. 8:1995;849-858 The paper shows a way to correct the database-derived potentials for the many-body effects and stresses the necessity of this correction.
72. Godzik A, Kolinski A, Skolnick J. Are proteins ideal mixtures of amino acids? Analysis of energy parameter set. of outstanding interest Protein Sci. 4:1995;2107-2117 This paper presents a very useful comparative analysis of various popular sets of mean-force two-body potentials that are based on protein structures. The authors show that the difference between these sets can be traced to the difference in a reference state used to define the zero energy. After a correction for this difference, various sets of potentials correlate at the level of 50-80%, although before the correction the correlation was much worse. In addition, the authors point to a good correlation of potentials derived from big and small proteins, and to an unexplained bad correlation between the potentials derived from crystallographic protein structures and those derived from NMR structure.
73. Jernigan R, Bahar I. Structure-derived potentials and protein simulations. of special interest Curr Opin Struct Biol. 6:1996;195-209 A critical review of knowledge-based mean-force potentials. The review stresses that a correct determination of a 'reference state' (e.g. 'residues dissolved in water' or 'residues dissolved in a random mixture of amino acids') is critically important for successful application of the potential functions.
74. Maiorov VN, Crippen GM. Contact potential that recognises the correct folding of globular proteins. J Mol Biol. 227:1992;876-888.
75. Koretke KK, Luthey-Schulten ZA, Wolynes PG. Self-consistently optimised statistical mechanical energy functions for sequence structure alignment. of outstanding interest Protein Sci. 5:1996;1043-1059 This paper describes an optimization of energy functions and their application to the discrimination of correctly folded protein chains from the misfolded ones. The main idea of the optimization [22] is that the optimal energy functions can be obtained by maximising the ratio of the stability gap, between the average energy of the correct fold and the mean energy of the full ensemble of misfolded structures, to the standard deviation of energies of the misfolded structures.
76. Abagyan R, Frishman D, Argos P. Recognition of distantly related proteins through energy calculations. Proteins. 19:1994;132-140.
77. Minor DL Jr, Kim PS. Context is a major determinant of β-sheet propensity. Nature. 371:1994;264-267.
78. Galzitskaya OV, Finkelstein AV. Folding of chains with random and edited sequences: similarities and differences. Protein Eng. 8:1995;883-892.
79. Angel E, Bellman R. Dynamic Programming and Partial Differential Equations. 1972;Academic Press, New York.
80. Waterman MS. . of special interest Introduction to Computational Biology. Maps, Sequences and Genomes. 1995;Chapman & Hall, London, An excellent, but not an easy, mathematical book for biologists that contains not only theorems but also numerous algorithms. For a reader of this review, the most relevant chapters are 9-12, which deals with dynamic-programming algorithms, pattern probabilities and statistics.
81. Needelman SB, Wunsch CD. A general method applicable to the search for similarities in the amino acid sequences of the proteins. J Mol Biol. 48:1971;443-453.
82. Sippl MJ, Flöckner H. Threading thrills and threats. Structure. 4:1996;15-19.
83. Finkelstein AV, Reva BA. Search for a stable fold of a short chain in a molecular field. Protein Eng. 5:1992;617-624.
84. Taylor WR, Orengo CA. Protein structure alignment. J Mol Biol. 208:1989;1-22.
85. Kubo R. Statistical Mechanics. 1965;North-Holland Publishing Company, Amsterdam.
86. Lifshitz IM, Grosberg AY, Khokhlov AR. Some problems of the statistical physics of polymer chains with volume interactions. Rev Mod Phys. 50:1979;683-713.
87. * is favourable for self-organization of native-like intermediates in protein folding.
88. 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. 239:1994;249-275.
89. Lee C. Predicting protein mutant energies by self-consistent ensemble optimisation. J Mol Biol. 236:1994;918-939.
90. Koehl P, Delarue M. A self-consistent mean-field approach to simultaneous gap closure and side-chain positioning in homology modelling. of outstanding interest Nat Struct Biol. 2:1995;163-170 This paper describes a new approach which simultaneously calculates loop conformation and sidechain positioning in homology modelling. It uses a database of possible loop conformations, a rotamer sidechain library, and an iterative refinement based on the SCMF theory.
91. Koehl P, Delarue M. Mean-field minimisation methods for biological macromolecules. of special interest Curr Opin Struct Biol. 6:1996;222-226 This review gives a simple description of SCMF theory and describes a number of applications of this theory to problems of protein-structure prediction, especially the prediction of the sidechain conformations within a given protein fold.
92. Laurents DV, Subbiah S, Levitt M. Different protein sequences can give rise to highly similar folds through different stabilising interactions. Protein Sci. 3:1994;1938-1944.
93. Bryant SH. Evaluation of threading specificity and accuracy. of special interest Proteins. 26:1996;172-185 A threading experiment using globins shows that a sequence can recognize, as a cognate, the fold of its remote homologue only when the native fold of this sequence and the fold of the homologue have at least 60% of super-imposible analogous residue sites. Not all the pairs of globins satisfy this condition. Threading alignment is even more sensitive to structure similarity (the alignment places the majority of residues correctly only when a deviation of superimposed-residue sites is less than 2 Å).
94. Chothia C. One thousand protein families for the molecular biologist. Nature. 357:1992;543-544.
95. Rost B. TOPITS: threading one-dimensional predictions into three-dimensional structure. of special interest ISMB-95: Proceedings of the 3rd Internatal Conference on Intelligent Systems for Molecular Biology. 1995;314-320 AAA Press, Menlo Park, The paper describes a simplified threading approach based the on alignments of predicted (using PHD [31]) secondary structures and predicted residue accessibilities with those observed in proteins of known 3D structure.
96. Hubbard JT, Park J. Fold recognition and ab initio structure prediction using hidden Markov models and -strand pair profiles. Proteins. 23:1995;398-402.
97. Ptitsyn OB. Structures of folding intermediates. of special interest Curr Opin Struct Biol. 5:1995;74-78 A review of kinetic and thermodynamic folding intermediates, which gives special attention to the native like highly-ordered molten globules, to the looser premolten globules, and to the barrier between them.
98. Ptitsyn OB. Molten globule and protein folding. of outstanding interest Adv Prot Chem. 47:1995;83-229 A comprehensive review describing the present state of studies of the molten globule state and its role in protein folding and in physiological processes.
99. Tanaka T, Kureda Y, Kimura H, Kidokoro S, Nakamura H. Cooperative deformation of a de novo designed protein. Protein Eng. 7:1994;969-976.
100. Yue K, Dill KA. Folding proteins with simple energy function and extensive conformational searching. of special interest Protein Sci. 5:1996;254-261 The described method uses very few energy parameters: one for hydrophobic interactions of nonpolar atoms; another for interaction of the hydrogen-bond donors and acceptors; and the hard-sphere van der Waals radii parameters. Amino acids are represented as 'united atoms', except for polar hydrogens, which are explicit. Each amino acid has between 3 (for proline) to 6 (for glycine) possible mainchain conformations. A new very fast 'constraint-based exhaustive' searching method is used to look through all the possible compact conformations. For the four small proteins examined, this method finds a relatively small number of low-energy conformations, among which are native-like folds. The rmsd of these native-like folds from the native ones is 4-5.
101. Reva BA, Sanner MF, Olson AJ, Finkelstein AV. Lattice modeling: accuracy of energy calculations. J Comput Chem. 17:1996;1025-1032.
102. Abagyan RA. Towards protein folding by global energy minimisation. FEBS Lett. 325:1993;17-22.
103. Bryngelson JD. When is a potential accurate enough for structure prediction? Theory and application to a random heteropolymer model of protein folding. J Chem Phys. 100:1994;6038-6045.
104. Pande VS, Grosberg AY, Tanaka T. How accurate must be potentials for successful modeling of protein folding. J Chem Phys. 103:1995;9482-9491.
105. De Araujo AFP, Pochepski TC. Monte Carlo simulations of protein folding using unexact potentials; how accurate must parameters be in order to preserve the essential features of energy landscape? of special interest Fold Des. 1:1996;299-314 This paper shows that a successful folding requires a sufficiently large energy gap between the lowest energy fold and the other ones.
106. Benner SA, Chelvanayagam G, Tarcotte M. Bona fide predictions of protein secondary structure using transparent analysis of multiple sequence alignments. Chem Rev. 1997;. in press.
107. Watson JD, Crick FHC. A structure of deoxyribose nucleic acid. Nature. 171:1953;737-738.
108. Ptitsyn OB. Protein as edited statistical copolymer? Srinivasan R, Sarma RM. Conformation in Biology. 1983;49-58 Academic Press, New York.
109. Sali A, Shakhnovich E, Karplus M. Kinetics of protein folding. A lattice model study of the requirements for folding to the native state. J Mol Biol. 235:1994;1614-1636.
110. Shakhnovich EI, Gutin AM. Formation of a unique structure in polypeptide chains. Theoretical investigation with the aid of replica approach. Biophys Chem. 34:1989;187-199.
111. Heinemann U, Hahn M. Circular permutation of protein sequence: not so rare? Trends Biochem Sci. 20:1995;349-350.
112. Finkelstein AV, Lobanov MY, Khlebalov ME, Rykunov DS. Prediction of 3D-folds of protein chains [abstract]. 12th International Biophysics Congress. 1996 August 11-16;. Amsterdam.
113. Kraulis PJ. MOLSCRIPT: A program to produce both detailed and schematic plots of protein structures. J Appl Cryst. 24:1991;946-950.