Gromacs Drug/Enzyme complex solvation tutorial

GROMACS Tutorial for Drug Enzyme Complex.

John E. Kerrigan, Ph.D. IST/Academic Computing Services RWJMS University of Medicine and Dentistry of NJ 675 Hoes Lane, RM SB11 Piscataway, NJ 08854 Phone: (732) 235-4473 Fax: (732) 235-5252 Email: [email protected] Webpage: http://www2.umdnj.edu/~kerrigje

Gromacs Drug/Enzyme complex solvation tutorial HEADER SERINE PROTEASE

TITLE COMPND COMPND COMPND COMPND SOURCE SOURCE SOURCE SOURCE KEYWDS EXPDTA AUTHOR REVDAT JRNL JRNL JRNL JRNL JRNL

26-NOV-97 1AZ8

BOVINE TRYPSIN COMPLEXED TO BIS-PHENYLAMIDINE INHIBITOR MOL_ID: 1; 2 MOLECULE: TRYPSIN; 3 CHAIN: NULL; 4 EC: 3.4.21.4 MOL_ID: 1; 2 ORGANISM_SCIENTIFIC: BOS TAURUS; 3 ORGANISM_COMMON: BOVINE; 4 ORGAN: PANCREAS HYDROLASE, SERINE PROTEINASE X-RAY DIFFRACTION R.ALEXANDER,A.SMALLWOOD 1 13-JAN-99 1AZ8 0 AUTH R.ALEXANDER,T.MADUSKUIE,K.MCNAMARA,A.SMALLWOOD, AUTH 2 A.WEI,P.STOUTEN TITL UNEXPECTED BINDING MODE OF A BIS-PHENYLAMIDINE TITL 2 FACTOR XA INHIBITOR COMPLEXED TO BOVINE TRYPSIN REF TO BE PUBLISHED
NH2 NH2

O H3CO NH2 NH2

Note: You will generate gromacs (*.gro) structure files in this tutorial. To view these files, you must use either gOpenMol (Download from: http://staff.csc.fi/~laaksone/gopenmol/gopenmol.html) or VMD (Download from: http://www.ks.uiuc.edu/Research/vmd/ ). Gromacs is a very powerful molecular simulation package.[1] Like other molecular mechanics/dynamics software packages, Gromacs uses an internal set of databases, which contain parameters for the amino acids, nucleic acids, cofactors, and some lipids that may be present in your PDB file. Gromacs does not know how to parameterize anything else in your PDB file (i.e. the HETATM records). You must provide these missing parameters. In this tutorial, you will learn how to use the PRODRG2 server to parameterize your HETATM group (the drug) and include it into your calculations. NOTE: This tutorial assumes that you are using Gromacs in a linux/unix environment. If you are using the Gromacs port to Windows, you will need to run the commands that follow from within the Windows command shell (formerly dosshell). Go To Start > Run and type cmd in the Run dialog to run the command shell.

Gromacs Drug/Enzyme complex solvation tutorial

Click OK

The Windows 2000 command shell. Establish a working directory for your project. For example, to make a project directory in your zip drive (assuming drive letter D: on some systems the zip drive is assigned drive letter E). This may also be accomplished using Windows Explorer (use the MS window key E key combination to launch Explorer). cd D: mkdir project cd project In linux/unix, establish a project directory. Use mkdir project (same command as in Windows). Then cd in to that directory and perform all of your work in your project directory. 3

Gromacs Drug/Enzyme complex solvation tutorial

Open 1AZ8.pdb in MOE. Open Sequence Editor and delete the water chain (chain #3) and the drug chain (IN4; chain #2). With your mouse click on chain #2 to select it. Go to Edit > Delete Selected Chains. Perform the same operation to delete chain #3.

Save the remaining protein as trp.pdb. Use File > Save As

File > Close in MOE File > Open > 1AZ8.pdb Now open the sequence editor and delete chains #1 and #3. Save the drug as drg.pdb. *********************************************************************** Next we will use the Dundee PRODRG[2] server to build a Gromacs 87 topology for the drug (drg.pdb).

Gromacs Drug/Enzyme complex solvation tutorial

In your browser, go to http://davapc1.bioch.dundee.ac.uk/programs/prodrg/prodrg.html and copy and paste (ctrl-v) your drg.pdb coordinates into the empty text box on the webpage. The view above is what you will see initially before pasting your coordinates. Check the following options: Chirality Full charges Energy Minimization Click Run PRODRG. 5 Yes Yes No

Gromacs Drug/Enzyme complex solvation tutorial

Download the Zipped Archive. The 00README file has the following information about the files contained in the archive:
This is an explanation of what the output files mean/contain. Not all files are listed, only the essential stuff. DRGCNS.PAR DRGCNS.TOP DRGDRG.MDS DRGDRG.TOP DRGFIN.GRO DRGFIN.PDB DRGGMX.TOP drggmx.itp) DRGHEX.TOP DRGML2.TOP DRGML2.TOPH DRGNOH.PDB DRGPOH.PDB DRGSHX.TOP DRGTRS.O DRGWIF.TOP : : : : : : : : : : : : : : : CNS parameter file CNS topology file the PRODRG MOLDES (see paper) the GROMOS87 topology building block final coordinates (in GROMOS87 format, not PDB!) after EM the final PDB file with all hydrogens added GROMACS .itp file (NOTE: This is now correctly named HEX topology MOL2 file with only polar hydrogens MOL2 file with all hydrogens the final PDB file without any hydrogens added the final PDB file with only polar hydrogens added SHELXL topology O torsion database (read in with: read DRGTRS.O in O) the WHAT IF topology

We want the DRGGMX.ITP file for building the topology for the drug. In addition, we will need the DRGFIN.GRO file for building the coordinate file (*.GRO). Extract the DRGPOH.PDB file for reference use. Rename the DRGGMX.ITP file to drg.itp Process the trp.pdb file with pdb2gmx. pdb2gmx f trp.pdb o trp.gro p trp.top Select the Gromacs forcefield (Type 0 then hit <Enter>) Next we must edit the drg.itp, trp.gro, and trp.top files! Edit the drg.itp file. You should always edit the drg.itp file to make sure that the atom types and the charges make sense. (NOTE: See the Gromacs manual for a description of gromacs atom types.) Here is the finished atoms section of our file:
[ atoms ] ; nr type resnr resid 1 NT 1 IN4 2 H 1 IN4 3 H 1 IN4 4 CB 1 IN4 atom N3 HAA HAB C20 cgnr charge 1 0.145 1 0.072 1 0.072 1 0.422

Gromacs Drug/Enzyme complex solvation tutorial

5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 NT H H CB CR61 CR61 CR61 CR61 CB CH1 CH2 C O OS CH3 CH2 CH2 CB CR61 CR61 CR61 CR61 CB CB NT H H NT H H 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 IN4 N4 HAF HAE C6 C5 C1 C2 C3 C4 C10 C14 C22 O3 O4 C8 C9 C11 C12 C19 C17 C7 C13 C15 C16 N1 HAH HAG N2 HAD HAC 1 1 1 2 2 2 2 3 3 3 3 4 4 4 4 4 5 5 6 6 6 6 6 7 7 7 7 7 7 7 0.145 0.072 0.072 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.469 -0.576 -0.113 0.089 0.131 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.423 0.144 0.072 0.072 0.145 0.072 0.072

No errors in atom typing were found. No charges were missing; however, always be sure to check for missing charges on heteroatoms. Notice that charges are localized into groups. The cgnr field is the charge group number. For comparison study, you should estimate charges using a quantum mechanics software package like Gaussian (www.gaussian.com) or Spartan (www.wavefun.com). The itp file contains molecule information only. We are building this itp file from a top file, and will need to remove some top file lines (already present in the top file generated by pdb2gmx). Open DRGFIN.GRO in a text editor (or view the contents in a separate unix shell using the more command). Open trp.gro in a text editor (Word, vi, jot, etc.). Copy and paste the coordinates from DRFIN.GRO onto the end of the trp.gro file. Change the number at the top or beginning of the file from 2099 to 2133 to correct the total number of atoms in the file. The drug will be residue #224 (after 223ASN in the trp.gro file). Change the residue name of the drug to 224IN4. Adjust the columns accordingly and renumber the atoms in a continuous fashion from the end of the original file (i.e. the first atom in the DRG should be atom # 2100, the second 2101, and so on). Save trp.gro (overwrite the old file). 7

Gromacs Drug/Enzyme complex solvation tutorial

Edit trp.top using any text editor. Add #include drg.itp after the forcefield parameter include statement. Under the [ molecules ] section, add IN4 under Compound and 1 under #mols. Setup the box editconf bt octahedron f trp.gro o trp.gro c d 0.5 What you have done in this command is specify that you want to use a truncated octahedral periodic box. The c flag centers the molecule(s) in the box and the d 0.5 flag sets the dimensions of the box based upon setting the box edge approx 0.5 nm from the molecule(s). Ideally you should set d at no less than 0.5 nm to an upper limit of 1.0 nm or more depending upon the system. CAUTION: We are using as small a value as possible for this simulation. Normally, you would use a larger spacing. genbox cp trp.gro cs spc216.gro o trp_b4ion.gro p trp.top The genbox command generates the water box based upon the dimensions/box type that you had specified using editconf. In the command above, you are using the SPC water model. [3] The genbox program will add the correct number of water molecules needed to solvate your box of given dimensions. We are naming our output file using b4ion. We will need to use genion later neutralize the charge in this system. Before we can use genion, we must generate a tpr file for input. Therefore, we will setup the energy minimization, add ions and redo the energy min setup. Use the em.mdp file. Look into the contents of this file. It specifies a steepest descents minimization to remove bad van der Waals contacts. [LOCATION OF CPP ON WINDOWS MACHINES: On most systems this location is C:\gromacs\bin\cpp] Content of em.mdp:
title cpp define constraints integrator dt nsteps nstlist ns_type rlist coulombtype rcoulomb rvdw fourierspacing fourier_nx fourier_ny fourier_nz pme_order ewald_rtol = drg_trp = /lib/cpp ; location of cpp on SGI = -DFLEX_SPC ; Use Fergusons Flexible water model [4] = none = steep = 0.002 ; ps ! = 500 = 10 = grid = 0.9 = PME ; Use particle-mesh ewald = 0.9 = 1.0 = 0.12 0 0 0 4 1e-5

= = = = =

Gromacs Drug/Enzyme complex solvation tutorial

optimize_fft = yes ; ; Energy minimizing stuff ; emtol = 1000.0 emstep = 0.01

grompp f em.mdp c trp_b4ion.gro p trp.top o trp_b4ion.tpr You will notice that there is one warning noting that the system has a nonzero total charge (+9.0). We must neutralize this charge with 9 Cl- ions (for PME total charge must = ~0). We will use a program called genion to add the Cl ions. genion s trp_b4ion.tpr o trp_b4em.gro nname Cl nn 9 g trp_ion.log You will be prompted to select a group of continuous water molecules. Select Group 13 (SOL). You must make the following additions/corrections to your trp.top file: Add the line #include ions.itp after the include statement for the forcefield or the drug. Subtract 9 from the total number of SOL molecules and add a line for 9 Cl molecules at the end of the trp.top file. We must include the drug (IN4) and the ions (Cl) in the temperature coupling schemes in the pr.mdp and md.mdp files later on. Run grompp to setup the energy minimization. grompp f em.mdp c trp_b4em.gro p trp.top o trp_em.tpr Run the energy minimization in background (Use the ampersand & at the end of the command). [WINDOWS USERS: You cannot run jobs in background in Windows. Omit the ampersand and nohup from the command below.] nohup mdrun s trp_em.tpr o trp_em.trr c trp_b4pr.gro g em.log e em.edr & nohup is used to prevent hang-ups of background jobs on linux/unix systems. Use the tail command to check on the progress of the minimization. Our run converged to 1000 in 335 steps. [WINDOWS USERS: You do not have the tail command. Use the type command instead. For example, use type md.log to view the contents of em.log.] tail 15 em.log Setup of the Position-Restrained Dynamics: 9

Gromacs Drug/Enzyme complex solvation tutorial We will run a position-restrained dynamics simulation to soak the water and the drug into the drug-enzyme complex. In this run, we will restrain the atom positions of the protein to restrict their movement in the simulation (i.e. the atom positions are restrained not fixed!). The water and the drug will be permitted to relax about the protein. The relaxation time of water is 10 ps. Therefore, we will use a 20 ps dynamics run to perform the soak. Under coulombtype, PME stands for Particle Mesh Ewald electrostatics. PME is the best method for computing longrange electrostatics (gives more reliable energy estimates). [5, 6] The all-bonds option under constraints applies the Linear Constraint algorithm[7] for fixing all bond lengths (important to use this option when dt > 0.001 ps). We are using Berendsens temperature and pressure coupling methods. [8]
title = trp_drg warnings = 10 cpp = /lib/cpp ; location of cpp on SGI define = -DPOSRES constraints = all-bonds integrator = md dt = 0.002 ; ps ! nsteps = 10000 ; total 20.0 ps. nstcomm = 1 nstxout = 250 ; output coordinates every 0.5 ps nstvout = 1000 ; output velocities every 2.0 ps nstfout = 0 nstlog = 10 nstenergy = 10 nstlist = 10 ns_type = grid rlist = 0.9 coulombtype = PME rcoulomb = 0.9 rvdw = 1.0 fourierspacing = 0.12 fourier_nx = 0 fourier_ny = 0 fourier_nz = 0 pme_order = 6 ewald_rtol = 1e-5 optimize_fft = yes ; Berendsen temperature coupling is on in four groups Tcoupl = berendsen tau_t = 0.1 0.1 0.1 0.1 tc_grps = protein sol IN4 Cl ref_t = 300 300 300 300 ; Pressure coupling is on Pcoupl = berendsen pcoupltype = isotropic tau_p = 0.5 compressibility = 4.5e-5 ref_p = 1.0 ; Generate velocites is on at 300 K. gen_vel = yes gen_temp = 300.0 gen_seed = 173529

10

Gromacs Drug/Enzyme complex solvation tutorial grompp f pr.mdp c trp_b4pr.gro p trp.top o trp_pr.tpr [WINDOWS USERS: You cannot run jobs in background in Windows. Omit the ampersand and nohup from the command below.] nohup mdrun s trp_pr.tpr o trp_pr.trr c trp_b4md.gro g pr.log e pr.edr & Setting up the Molecular Dynamics Simulation: The parameters shown below are for the Gromacs 87 force field, which we are using in this example. Consult the Gromacs manual for optimal settings for other force fields (e.g. the GROMOS 96 force field uses nstlist = 5 and rvdw = 1.4 is required). Note: nstxout sets the snapshot frequency (e.g. nstxout = 500 writes out coordinates every 500 steps or 1.0 ps). Content of md.mdp for NPT ensemble 50 ps simulation.
title = trp_drg MD cpp = /lib/cpp ; location of cpp on SGI constraints = all-bonds integrator = md dt = 0.002 ; ps ! nsteps = 25000 ; total 50 ps. nstcomm = 1 nstxout = 500 ; output coordinates every 1.0 ps nstvout = 0 nstfout = 0 nstlist = 10 ns_type = grid rlist = 0.9 coulombtype = PME rcoulomb = 0.9 rvdw = 1.0 fourierspacing = 0.12 fourier_nx = 0 fourier_ny = 0 fourier_nz = 0 pme_order = 6 ewald_rtol = 1e-5 optimize_fft = yes ; Berendsen temperature coupling is on in four groups Tcoupl = berendsen tau_t = 0.1 0.1 0.1 0.1 tc-grps = protein IN4 sol Cl ref_t = 300 300 300 300 ; Pressure coupling is on Pcoupl = berendsen pcoupltype = isotropic tau_p = 0.5 compressibility = 4.5e-5 ref_p = 1.0 ; Generate velocites is on at 300 K. gen_vel = yes

11

Gromacs Drug/Enzyme complex solvation tutorial

gen_temp gen_seed = = 300.0 173529

grompp f md.mdp c trp_b4md.gro p trp.top o trp_md.tpr [WINDOWS USERS: You cannot run jobs in background in Windows. Omit the ampersand and nohup from the command below.] nohup mdrun s trp_md.tpr o trp_md.trr c trp_pmd.gro g md.log e md.edr & This run will take a long time (2 to 3 hours depending on the speed of your processor). Use the tail command to check the md.log file. Or, if you are using the Windows command shell, use the type command. When the simulation is finished, compress the trajectory using trjconv. trjconv f filename.trr o filename.xtc Remove the *.trr file to save on disk space (use rm trp_md.trr). Use ngmx to view the trajectory (you may also download to your PC and use VMD to view the trajectory. Note: VMD will ask for the trajectory file and the *.gro file for input.). (NOTE: ngmx does not run in Windows.) ngmx f trp_md.trr (or trp_md.xtc) s trp_md.tpr Select the Protein and the DRG groups. Use g_energy to analyze the potential energy output (the md.edr file). g_energy f md.edr o pe.xvg You will be prompted to select the energy term. Enter 10 (for potential) followed by 0 to end selection. Use the g_energy command to obtain other components (e.g. temperature, pressure, kinetic energy, etc.) Here is a plot of the potential energy data from our run. We performed a 50 ps dynamics simulation.

12

Gromacs Drug/Enzyme complex solvation tutorial

Tryptophan-IN4 Complex Gromacs FF

-196500 0 10 20 30 40 50

Potential Energy (kJ/mol)

-197000

-197500

-198000

-198500

-199000

Time (ps)

The potential energy equilibrates about an average of ~ 198100 kJ/mol.

Temperature Fluctuation
306 304

Temperature (K)

302 300 298 296 294 292 0 10 20 30 40 50

Time (ps)

The temperature plot shows the normal oscillation behavior of the temperature about the desired average (300 K).

13

Gromacs Drug/Enzyme complex solvation tutorial

Pressure Fluctuation
400 300 200

Pressure (bar)

100 0 0 -100 -200 -300 -400 10 20 30 40 50

Time (ps)

Use g_hbond to analyze the number of hydrogen bonds between the drug (DRG) and the enzyme (Protein). Be sure to select the Protein and the DRG groups when prompted by the program. g_hbond f trp_md.xtc s trp_md.tpr num trp_hnum.xvg Select Group 12 (IN4) and Group 1 (Protein). g_hbond uses a default distance cutoff of 2.5 angstroms (0.25 nm) and a default angle cutoff of 60 degrees (i.e. bond angle must be 60 degrees or greater). The angle cutoff is rather liberal. Some researchers prefer a more stringent angle cutoff of 120 degrees. We used the default in the example above.

14

Gromacs Drug/Enzyme complex solvation tutorial

Hydrogen Bonds
8 7

No. of Hydrogen Bonds

6 5 4 3 2 1 0 0 10 20 30 40 50

Time (ps)

Use g_rms to obtain an RMSD plot of the protein backbone and the drug (IN4) throughout the simulation. Do the Backbone first g_rms s trp_md.tpr f trp_md.trr (or xtc) o bkbone_rmsd.xvg You will be prompted to select a Group. Enter 4 (for backbone). Compare to 1 Group in the reference. Enter 1 then Enter 4 for backbone again.
Protein Backbone RMSD
0.12 0.1 0.08 0.06 0.04 0.02 0 0 10 20 30 40 50

RMSD (nm)

Time (ps)

15

Gromacs Drug/Enzyme complex solvation tutorial Now run the g_rms program to obtain an RMS plot for the drug (IN4). Select Group 12 (IN4) in this analysis. g_rms s trp_md.tpr f trp_md.trr (or xtc) o drg_rmsd.xvg
Drug Positional RMSD
0.12 0.1 0.08 0.06 0.04 0.02 0 0 10 20 30 40 50

RMSD (nm)

Time (ps)

The backbone RMSD indicates that the rigid protein structure equilibrates rather quickly in this simulation (after 20 ps). The drug does not equilibrate until after 30 ps. The RMSD for the drug is more variable indicative of its mobility within the binding pocket. Compute an average structure based upon the equilibration of the drug. We will use the g_rmsf program. g_rmsf f trp_md.trr s trp_md.tpr b 30 e 50 ox trp_avg.pdb o trp_av.xvg When prompted to select a Group for the calculation, use the System group (Enter 0). Refinement of the average structure. Average structures tend to be very crude and require energy minimization. We will perform an in vacuo refinement of our structure. Our first step will require that we remove all of the solvent (water) and ions (Cl) in our average structure (use a text editor). Separate out the protein (tryspin) into a separate file. Run pdb2gmx on the protein. Use editconf to convert the drg.pdb to a gro file. Append the drg.gro coordinates back to trp.gro as you did before. Use the drg.itp file that you generated previously in your top file. Run Steepest Descents followed by conjugate gradient using the mdp files that follow.

16

Gromacs Drug/Enzyme complex solvation tutorial

; Steepest Descents in vacuo ; User kerrigan (236) ; ; Input file ; cpp = /usr/bin/cpp ; location of cpp on linux constraints = none integrator = steep nsteps = 400 ; ; Energy minimizing stuff ; emtol = 1000 emstep = 0.01 nstcomm ns_type morse coulombtype vdw_type rlist rcoulomb rvdw rcoulomb_switch rvdw_switch epsilon_r Tcoupl Pcoupl gen_vel = = = = = = = = = = = = = = 1 grid no Shift Shift 1.0 1.2 1.2 1.0 1.0 6.0 no no no

; Conjugate gradient with morse potential ; in vacuo ; User kerrigan (236) ; ; Input file ; cpp = /usr/bin/cpp ; location of cpp on linux constraints = none integrator = cg nsteps = 3000 ; ; Energy minimizing stuff ; emtol = 100 emstep = 0.01 nstcgsteep = 1000 nstcomm morse coulombtype vdw_type ns_type rlist rcoulomb rvdw rcoulomb_switch = = = = = = = = = 1 yes Shift Shift grid 1.0 1.2 1.2 1.0

17

Gromacs Drug/Enzyme complex solvation tutorial

rvdw_switch epsilon_r Tcoupl Pcoupl gen_vel = = = = = 1.0 6.0 no no no

In analyzing a drug/enzyme complex, observations are key. You must ask the following questions: Was the complex stable to the simulation conditions? (i.e. Did the drug remain in the active site pocket or did it fall out?) Upon equilibration did the complex become more stable? Why? What factors contributed to the stability? (hydrogen bonds; hydrophobic pockets; water bridges or other solvent interaction) Using VMD to view the trajectory. Open VMD. [9] In VMD Main, go to File > New Molecule

Click on Browse and select trp_pmd.gro; then, click on Load

In VMD Main, select File > New Molecule and select trp_md.xtc (or trr) to load the trajectory. Make sure that the trp_pmd.gro file is highlighted in VMD Main before loading the trajectory. In VMD Main select Display > Orthographic

18

Gromacs Drug/Enzyme complex solvation tutorial Use the mouse to rotate (default mode is rotate). To Zoom or Scale, hit the s key to place the mouse in scale mode. Hit the t key to use translate mode. Hit the r key to go back to rotate mode. In VMD Main, Open Graphics > Representations

Under selected atoms, type protein in place of all and hit enter (the water goes away and so does your drug). Click on Create Rep to create an additional representation. In the selected atoms box enter resname IN4. Under Drawing Method, select Licorice. Close the Graphics Representations dialog. In VMD Main, move the speed bar to the midway point and click on the arrow in the lower right-hand corner. This will replay the trajectory over and over again. Note the movement of the drug relative to the protein. Click on the same arrow again to stop the replay. Rotate the structure until you have a nice view of the drug. Open Graphics Representations again. Select the protein representation. Change Drawing Method to Surf and Coloring Method to Charge. The surface will be colored entirely in green because our input does not have partial atomic charges. Nevertheless, we have a solvent

19

Gromacs Drug/Enzyme complex solvation tutorial accessible surface on the receptor illustrating the snug fit of the drug in the binding pocket. VMD uses the Surf program that was developed at UNC-Chapel Hill. [10]

Bibliography 1. 2. Lindahl, E., B. Hess, and D. van der Spoel, GROMACS 3.0: a package for molecular simulation and trajectory analysis. J. Mol. Model, 2001. 7: p. 306-317. van Aalten, D.M.F., R. Bywater, J.B.C. Findlay, M. Hendlich, R.W.W. Hooft, and G. Vriend, PRODRG, a program for generating molecular toppologies and unique molecular descriptors from coordinates of small molecules. J. Comput. Aided Mol. Design, 1996. 10: p. 255-262. Berendsen, H.J., J.P. Postma, W.F. van Gunsteren, and J. Hermans, Interaction models for water in relation to protein hydration., in Intermolecular Forces, B. Pullman, Editor. 1981, D. Reidel Publishing Co.: Dordrecht. p. 331-342. Ferguson, D., Parametrization and evaluation of a flexible water model. J. Comput. Chem., 1995. 16: p. 501-511. Darden, T., D. York, and L. Pedersen, Particle Mesh Ewald: An N-log(N) method for Ewald sums in large systems. J. Chem. Phys., 1993. 98: p. 10089-10092. Essmann, U., L. Perera, M.L. Berkowitz, T. Darden, H. Lee, and L. Pedersen, A smooth particle mesh ewald potential. J. Chem. Phys., 1995. 103: p. 8577-8592. Hess, B., H. Bekker, H. Berendsen, and J. Fraaije, LINCS: A Linear Constraint Solver for molecular simulations. J. Comp. Chem., 1997. 18: p. 1463-1472.

3. 4. 5. 6. 7.

20

Gromacs Drug/Enzyme complex solvation tutorial 8. 9. 10. Berendsen, H.J.C., J.P.M. Postma, W.F. vanGunsteren, A. DiNola, and J.R. Haak, Molecular dynamics with coupling to an external bath. J. Chem. Phys., 1984. 81(8): p. 3584-3590. Humphrey, W., A. Dalke, and K. Schulten, VMD - Visual Molecular Dynamics. J. Molec. Graphics, 1996. 14.1: p. 33-38. Varshney, A., F.P. Brooks, and W.V. Wright, Linearly Scalable Computation of Smooth Molecular Surfaces. IEEE Comp. Graphics and Applications, 1994. 14: p. 19-25.

21