Evaluating protein models

The process of determining the best alignment and how to construct, refine, and evaluate protein models is less than a precise science and more... 

Abstract. Molecular dynamics (MD) simulations of proteins provide descriptions of atomic motions, which allow to relate observable properties of proteins to microscopic processes. Unfortunately, such MD simulations require an enormous amount of computer time and, therefore, are limited to time scales of nanoseconds. We describe first a fast multiple time step structure adapted multipole method (FA-MUSAMM) to speed up the evaluation of the computationally most demanding Coulomb interactions in solvated protein models, secondly an application of this method aiming at a microscopic understanding of single molecule atomic force microscopy experiments, and, thirdly, a new method to predict slow conformational motions at microsecond time scales. 

L Chiche, LM Gregoret, FE Cohen, PA Kollman. Protein model structure evaluation using the solvation free energy of folding. Proc Natl Acad Sci USA 87 3240-3244, 1990. 

L Holm, C Sander. Evaluation of protein models by atomic solvation preference. J Mol Biol 225 93-105, 1992. 

Figure 2-7. Origins of the increased O2 binding energy in IPNS when the protein is included in an ONIOM model. (A) A comparison of the optimized geometries from an active-site model (silver) and an ONIOM protein model (dark grey), show that the artificial structural relaxation of the active-site model is more pronounced for the reactant state than for the product state. (B) Contributions to O2 binding from the surrounding protein, evaluated only at the MM level (Adapted from Lundberg and Morokuma [26], Reprinted with permission. Copyright 2007 American Chemical Society.)...

Alwyn Jones, T. and Kleywegt, G. J. (1999) CASP3 comparative modeling evaluation. Proteins Suppl. 3, 30-46. 

For the orientation search (often called a rotation search), the computer is looking for large values of the model Patterson function Pmodel( ,v,w) at locations corresponding to peaks in the Patterson map of the desired protein. A powerful and sensitive way to evaluate the model Patterson is to compute the minimum value of Pmodel(w,v,w) at all locations of peaks in the Patterson map of the desired protein. A value of zero for this minimum means that the trial orientation has no peak in at least one location where the desired protein exhibits a peak. A high value for this minimum means that the trial orientation has peaks at all locations of peaks in the Patterson map of the desired protein. 

A. Kolinski, J.M. Bujnicki, Generalized protein structure prediction based on combination of fold-recognition with de novo folding and evaluation of models. Proteins 61(S7), 84-97 (2005)... 

Originated from physics, the potential of mean force was first proposed by Tanaka and Scheraga to evaluate protein structure models [37] and substantially extended by Miyazawa and Jernigan, Sippl, and others [38,39,42-44],... 

Venclovas C, Zemla A, Fidehs K, Moult J. Criteria for evaluating protein structures derived from comparative modeling. Proteins 1997 29(Suppl. 1) 7-13. 

The 3D models are generally evaluated by relying on geometrical preferences of the amino acid residues or atoms that are derived from known protein structures. Essential for interpreting 3D protein models is the estimation of their accuracy, both the overall accuracy and the accuracy in the individual regions of a model. The errors in models arise from two main sources, the failure of the conformational search to find the optimal conformation and the failure of the scoring function to identify the optimal conformation. 

The properties of three proteins, cytochrome-c, carboxypeptidase and bovine albumin were used to evaluate the model. The molecular weight and diffusion coefficient of each protein is listed in Table 1 (22). These molecular weight values represent the molecular weight range typical of many proteins. The effect of various operating and column parameters on the band broadening for each of these proteins as well as the resolution between proteins was investigated. 

Sali A, Potterton L, Yuan F, van Vlijmen H, Karplus M. Evaluation of comparative protein modeling by MODELLER. Proteins Struct Funct Genet 1995 23 318-326. 

We have developed such a force field, named SPEEDY (Simplified Parameters for Energy Evaluation and DYnamics). SPEEDY uses the QPACK protein model and residue interaction potentials.- The terms of the potential function are analogous to those used in molecular mechanics calculations. Since the backbone virtual dihedral and bond angles are dependent on the complicated geometry of the polypeptide backbone, these terms are not represented independently. Residue interactions are represented by Lennard-Jones type functions scaled by their... 

L. Chiche, L. M. Gregoret, R. E. Cohen, and P. A. Kollman, Proc. Natl. Acad. Sci. U.S.A., 87, 3240 (1990). Protein Model Structure Evaluation Using the Solvation Free Energy of Folding. 

---