Improving physical realism, stereochemistry, and side-chain accuracy in homology modeling: Four approaches that performed well in CASP8

Abstract

A correct alignment is an essential requirement in homology modeling. Yet in order to bridge the structural gap between template and target, which may not only involve loop rearrangements, but also shifts of secondary structure elements and repacking of core residues, high-resolution refinement methods with full atomic details are needed. Here, we describe four approaches that address this "last mile of the protein folding problem" and have performed well during CASP8, yielding physically realistic models: YASARA, which runs molecular dynamics simulations of models in explicit solvent, using a new partly knowledge-based all atom force field derived from Amber, whose parameters have been optimized to minimize the damage done to protein crystal structures. The LEE-SERVER, which makes extensive use of conformational space annealing to create alignments, to help Modeller build physically realistic models while satisfying input restraints from templates and CHARMM stereochemistry, and to remodel the side-chains. ROSETTA, whose high resolution refinement protocol combines a physically realistic all atom force field with Monte Carlo minimization to allow the large conformational space to be sampled quickly. And finally UNDERTAKER, which creates a pool of candidate models from various templates and then optimizes them with an adaptive genetic algorithm, using a primarily empirical cost function that does not include bond angle, bond length, or other physics-like terms.

Figure 1

Figure 1A. The 1D knowledge-based potential of the backbone Phi dihedral of Arg. To aid visualization, the energy axis is flipped: favorable low energy regions are at the top and colored yellow, high energy regions are colored blue. The red arrow indicates the current conformation shown in ‘D’. B: The 2D PhiPsi potential, most famous for its relationship with the Ramachandran plot. C: The 3D PhiPsiChi1 potential, the highest energy regions are transparent for clarity. D: Distribution of knowledge-based potentials, using the most complicated case of four Chi dihedrals as an example. A residue like Arg (or Lys) has seven 1D potentials (smallest rings at the top: Omega (not shown), Phi, Psi, Chi1–4), seven 2D potentials (rings in the middle: Psi-1Phi, PhiPsi, PhiChi1, PsiChi1, Chi12, Chi23, Chi34) and three 3D potentials (at the bottom: PhiPsiChi1, Chi123, Chi234). Note that the thickness of the rings symbolizes the weight: The 1D Chi4 potential contributes more because Chi4 is only covered by one 2D and one 3D potential. Likewise, the 2D PhiPsi potential (magenta) accounts for the fact that Phi and Psi are covered by just one 3D potential. The 2D PhiChi1 and PsiChi1 potentials count only half to ensure that Chi1 is not overweighted. Finally, the Psi-1 Phi potential, which spans two subsequent residues, fills up the remaining gaps. Since this potential is special (the two dihedrals are not adjacent but separated by Omega), its weight was optimized independently (parameter 41 in Table 1). Graphics created with YASARA and Po-vRay.