A novel exhaustive search algorithm for predicting the conformation of polypeptide segments in proteins

Abstract

We present a fast ab initio method for the prediction of local conformations in proteins. The program, PETRA, selects polypeptide fragments from a computer-generated database (APD) encoding all possible peptide fragments up to twelve amino acids long. Each fragment is defined by a representative set of eight straight phi/psi pairs, obtained iteratively from a trial set by calculating how fragments generated from them represent the protein databank (PDB). Ninety-six percent (96%) of length five fragments in crystal structures, with a resolution better than 1.5 A and less than 25% identity, have a conformer in the database with less than 1 A root-mean-square deviation (rmsd). In order to select segments from APD, PETRA uses a set of simple rule-based filters, thus reducing the number of potential conformations to a manageable total. This reduced set is scored and sorted using rmsd fit to the anchor regions and a knowledge-based energy function dependent on the sequence to be modelled. The best scoring fragments can then be optimized by minimization of contact potentials and rmsd fit to the core model. The quality of the prediction made by PETRA is evaluated by calculating both the differences in rmsd and backbone torsion angles between the final model and the native fragment. The average rmsd ranges from 1.4 A for three residue loops to 3.9 A for eight residue loops.

Ab initio construction of polypeptide fragments: Accuracy of loop decoy discrimination by an all‐atom statistical potential and the AMBER force field with the Generalized Born solvation model

“…discrete states, with as few as four to as many as 55 states. 18,20,45,46 In light of the strong correlation between the size of a state set and its ability to accurately represent native protein structures, 46 we use residue-specific, propensity-weighted state sets with as many as 72 2 states per residue. The state sets are derived from the Top500 database of protein structures, after eliminating residues where any main-chain atom has B-factor Ͼ 30.0 Å , 2 van der Waals overlap Ͼ 0.4 Å , 44 or grossly incorrect geometry.…”

Exploring the energy landscapes of flexible molecular loops using higher‐dimensional continuation

“…The topology of the conformational space seems to be a Klein bottle, but futher analysis is required to confirm this point [39]. The minimum root mean square error (RMSE) of the charts centers with respect to the crystal structure in the PDB considering all the atoms in the loop, including side chains, is about 0.2Å, which is comparable to the error obtained with other methods [23,17]. The conformation in the atlas giving the minimum RMSE is the one with minimum energy all over the conformational space at the given resolution.…”

“…Finally, we consider the loop formed by the residues 120 to 124 of the dihydrofolate reductase of the lactobacillus casei (PDB code 3DFR). This loop that has been analyzed before [23,17], is used here to demonstrate the applicability of the proposed techniques to problems with conformational spaces of higher dimensionality. In all cases, the initial point to start the analysis is obtained with a Jacobian pseudo-inverse numerical inverse kinematics process [59].…”

MOLS sampling and its applications in structural biophysics

“…The computational prediction of the three-dimensional structure of the loops has received extensive specific attention, and several algorithms have been developed (Sudarsanam et al 1995;van Vlijmen and Karplus 1997;Wojcik et al 1999;Fiser et al 2000;DePristo et al 2003;Deane and Blundell 2000). There have also been several attempts to classify loop structures (Leszczynski and Rose 1986;Ring et al 1992;Efimov 1993).…”