Modeling of loops in protein structures

Abstract

Comparative protein structure prediction is limited mostly by the errors in alignment and loop modeling. We describe here a new automated modeling technique that significantly improves the accuracy of loop predictions in protein structures. The positions of all nonhydrogen atoms of the loop are optimized in a fixed environment with respect to a pseudo energy function. The energy is a sum of many spatial restraints that include the bond length, bond angle, and improper dihedral angle terms from the CHARMM-22 force field, statistical preferences for the main-chain and side-chain dihedral angles, and statistical preferences for nonbonded atomic contacts that depend on the two atom types, their distance through space, and separation in sequence. The energy function is optimized with the method of conjugate gradients combined with molecular dynamics and simulated annealing. Typically, the predicted loop conformation corresponds to the lowest energy conformation among 500 independent optimizations. Predictions were made for 40 loops of known structure at each length from 1 to 14 residues. The accuracy of loop predictions is evaluated as a function of thoroughness of conformational sampling, loop length, and structural properties of native loops. When accuracy is measured by local superposition of the model on the native loop, 100, 90, and 30% of 4-, 8-, and 12-residue loop predictions, respectively, had Ͻ2 Å RMSD error for the mainchain N, C a , C, and O atoms; the average accuracies were 0.59 6 0.05, 1.16 6 0.10, and 2.61 6 0.16 Å, respectively. To simulate real comparative modeling problems, the method was also evaluated by predicting loops of known structure in only approximately correct environments with errors typical of comparative modeling without misalignment. When the RMSD distortion of the main-chain stem atoms is 2.5 Å, the average loop prediction error increased by 180, 25, and 3% for 4-, 8-, and 12-residue loops, respectively. The accuracy of the lowest energy prediction for a given loop can be estimated from the structural variability among a number of low energy predictions. The relative value of the present method is gauged by~1! comparing it with one of the most successful previously described methods, and~2! describing its accuracy in recent blind predictions of protein structure. Finally, it is shown that the average accuracy of prediction is limited primarily by the accuracy of the energy function rather than by the extent of conformational sampling.

Evolution and Physics in Comparative Protein Structure Modeling

“…12,13 A third group of methods uses either distance geometry or optimization techniques to satisfy spatial restraints obtained from the sequence-template alignment. [14][15][16] There are also many methods that specialize in the modeling of loops [17][18][19] and side chains 20,21 within the restrained environment provided by the rest of the structure.…”

“…Second, all the restraints are combined into an objective function that is optimized to obtain the final model. of spatial restraints 15,17 (Figure 3). It is implemented in the computer program MODELLER, which is freely available to academic researchers via the web at http://guitar.rockefeller.edu.…”

“…They are used because they were found to result in more accurate models than the corresponding terms from a molecular mechanics force field. 17 The third type of a restraint is obtained from the molecular mechanics force field of CHARMM-22 37 and includes restraints on chemical bonds, angles, and improper dihedral angles. These molecular mechanics restraints enforce proper stereochemistry of the model.…”

“…Achieving a sufficient accuracy and a productive balance between these two types of a restraint is highly nontrivial, as evidenced by the difficulty of refining comparative protein structure models. 15,42 The emphasis on the objective function is justified because there is strong anecdotal evidence that the prediction accuracy is often limited by the accuracy of the energy function 15,17 and possibly by the accuracy of the corresponding protein and solvent representations. Thus, in the next three paragraphs, we comment on the current accuracy of the terms that describe local stereochemistry, nonbonded interactions between protein atoms, and nonbonded interactions between protein and solvent atoms (i.e., solvation effect).…”

Protein structure prediction based on fragment assembly and parameter optimization

“…The information on the tertiary structure of a protein is quite crucial in understanding the function and biological role of the protein. Popular approaches to this problem include comparative modeling [1][2][3][4][5] and fold recognition [6][7][8][9], which are classified as knowledge-based methods [5,10,11]. These methods use statistical information on sequences and their three-dimensional structures, in structural databases such as Protein Data Bank (PDB), to predict the unknown structure of a protein.…”