Numerical Errors and Chaotic Behavior in Docking Simulations

This work examines the sensitivity of docking programs to tiny changes in ligand input files. The results show that nearly identical ligand input structures can produce dramatically different top-scoring docked poses. Even changing the atom order in a ligand input file can produce significantly different poses and scores. In well-behaved cases the docking variations are small and follow a normal distribution around a central pose and score, but in many cases the variations are large and reflect wildly different top scores and binding modes. The docking variations are characterized by statistical methods, and the sensitivity of high-throughput and more precise docking methods are compared. The results demonstrate that part of docking variation is due to numerical sensitivity and potentially chaotic effects in current docking algorithms and not solely due to incomplete ligand conformation and pose searching. These results have major implications for the way docking is currently used for pose prediction, ranking, and virtual screening.

[1]  Kai Diethelm,et al.  The Limits of Reproducibility in Numerical Simulation , 2012, Computing in Science & Engineering.

[2]  R. Unger,et al.  Chaos in protein dynamics , 1997, Proteins.

[3]  Paul N. Mortenson,et al.  Diverse, high-quality test set for the validation of protein-ligand docking performance. , 2007, Journal of medicinal chemistry.

[4]  James A. Yorke,et al.  Using fractal to solve the multiple minima problem in molecular mechanics calculation , 2000, J. Comput. Chem..

[5]  K. Schulten,et al.  Difficulties with multiple time stepping and fast multipole algorithm in molecular dynamics , 1997 .

[6]  Robert D. Skeel,et al.  Dangers of multiple time step methods , 1993 .

[7]  T. Halgren.  Merck molecular force field. I. Basis, form, scope, parameterization, and performance of MMFF94 , 1996, J. Comput. Chem..

[8]  Miklos Feher,et al.  The effect of numerical error on the reproducibility of molecular geometry optimizations , 2008, J. Comput. Aided Mol. Des..

[9]  Miklos Feher,et al.  Fuzzy Clustering as a Means of Selecting Representative Conformers and Molecular Alignments , 2003, J. Chem. Inf. Comput. Sci..

[10]  Miklos Feher,et al.  Reducing Docking Score Variations Arising from Input Differences , 2010, J. Chem. Inf. Model..

[11]  Chris H. Q. Ding,et al.  Using Accurate Arithmetics to Improve Numerical Reproducibility and Stability in Parallel Applications , 2000, ICS '00.

[12]  J. Yorke,et al.  Fractal basin boundaries , 1985 .

[13]  T. Schlick,et al.  Extrapolation versus impulse in multiple-timestepping schemes. II. Linear analysis and applications to Newtonian and Langevin dynamics , 1998 .

[14]  T. Halgren.  MMFF VI. MMFF94s option for energy minimization studies , 1999, J. Comput. Chem..

[15]  Miklos Feher,et al.  Identifying potential binding modes and explaining partitioning behavior using flexible alignments and multidimensional scaling , 2001, J. Comput. Aided Mol. Des..

[16]  Kenji Onodera,et al.  Evaluations of Molecular Docking Programs for Virtual Screening , 2007, J. Chem. Inf. Model..

[17]  T. Schlick,et al.  Masking Resonance Artifacts in Force-Splitting Methods for Biomolecular Simulations by Extrapolative Langevin Dynamics , 1999 .

[18]  Miklos Feher,et al.  Effect of Input Differences on the Results of Docking Calculations , 2009, J. Chem. Inf. Model..