Molecular dynamics simulation of wild type, non-cleavable form and mutants were performed with Gromacs 5.0.2 software under Amber ff99SB-LIDN forcefield. Each initial structure was centered in a cubic box with 1.0 nm distance from the box edge. Periodic boundary conditions were applied to avoid artificial effects due to finite size of simulation box. The box was then filled with appropriate amount of TIP3P water molecules. System was made electrically neutral by replacing water molecules with required number of counter-ions using the genion tool of Gromacs package ( 31 (link)
). Energy minimization was done using steepest descent algorithm. The energy-minimized system was equilibrated by the position restrained simulation under canonical ensemble (NVT) followed by isothermal isobaric ensemble (NPT) in 500 and 1000 ps respectively. The initial velocities were derived from Maxwell-Boltzman distribution at 300 k. Productive unrestrained MD simulations were carried out in the NPT ensemble for 20 ns and bond lengths constrained with LINCS algorithm. The leapfrog algorithm with 2 fs time step was applied to integrate the equation of motion. Conditions of constant pressure 1 bar and temperature 300 K were accomplished by application of the Parinello-Rahman barostat and Nosé-Hoover thermostat. The particle mesh Ewald (PME) method was used for calculating the long-range electrostatics interactions. The utilities of Gromacs 5.0.2 such as gmx rms, gmx rmsf, gmx gyrate, gmx sasa, gmx density and gmx hbond were used to analyze MD simulation trajectories ( 32 (link)
, 33 (link)
). The program DSSP was used to study the secondary structure fluctuations ( 34 (link)
). Principal components analysis (PCA) reduces the dimensionality of the data obtained from MD simulations. PCA is described in detailed by Giuliani ( 35 (link)
). Here, it was performed to examine the protein global motions. The Gromacs utility “gmx covar” was used to calculate and diagonalize the covariance matrix of Cα atomic positions from the 20 ns trajectories of the wild type r-PA and its mutants. Then, the eigenvalues and corresponding eigenvectors of the matrix were produced. The gmx anaeig was used for analysis and plotting the eigenvectors ( 35 (link)
, 36 (link)
). All graphs were plotted using Xmgrace ( 37
). The gmx sham was used to calculate the free energy (G) by the first two eigenvectors extracted after PCA. 2D and 3D representation of the free energy landscape (FEL)
were plotted using GNUplot 5.2 (www.gnuplot.info ). The dynamic cross-correlation matrix (DCCM) that shows the fluctuations of the fluctuations between any two pair of Cα atoms
were calculated by the Bio3D package.
). Energy minimization was done using steepest descent algorithm. The energy-minimized system was equilibrated by the position restrained simulation under canonical ensemble (NVT) followed by isothermal isobaric ensemble (NPT) in 500 and 1000 ps respectively. The initial velocities were derived from Maxwell-Boltzman distribution at 300 k. Productive unrestrained MD simulations were carried out in the NPT ensemble for 20 ns and bond lengths constrained with LINCS algorithm. The leapfrog algorithm with 2 fs time step was applied to integrate the equation of motion. Conditions of constant pressure 1 bar and temperature 300 K were accomplished by application of the Parinello-Rahman barostat and Nosé-Hoover thermostat. The particle mesh Ewald (PME) method was used for calculating the long-range electrostatics interactions. The utilities of Gromacs 5.0.2 such as gmx rms, gmx rmsf, gmx gyrate, gmx sasa, gmx density and gmx hbond were used to analyze MD simulation trajectories ( 32 (link)
, 33 (link)
). The program DSSP was used to study the secondary structure fluctuations ( 34 (link)
). Principal components analysis (PCA) reduces the dimensionality of the data obtained from MD simulations. PCA is described in detailed by Giuliani ( 35 (link)
). Here, it was performed to examine the protein global motions. The Gromacs utility “gmx covar” was used to calculate and diagonalize the covariance matrix of Cα atomic positions from the 20 ns trajectories of the wild type r-PA and its mutants. Then, the eigenvalues and corresponding eigenvectors of the matrix were produced. The gmx anaeig was used for analysis and plotting the eigenvectors ( 35 (link)
, 36 (link)
). All graphs were plotted using Xmgrace ( 37
). The gmx sham was used to calculate the free energy (G) by the first two eigenvectors extracted after PCA. 2D and 3D representation of the free energy landscape (FEL)
were plotted using GNUplot 5.2 (
were calculated by the Bio3D package.
