|
| 1 | +After extracting both path.py and func.py to the same folder, the PATH program |
| 2 | +can be run using the command |
| 3 | + |
| 4 | +path.py -h |
| 5 | + |
| 6 | +This gives a list of all the flags for customizing the simulation. |
| 7 | + |
| 8 | +The program can be used in two different modes - "path" and "rock" mode. "path" |
| 9 | +mode calculates the most probable pathways between two equilibrium states of a |
| 10 | +protein. "rock" mode calculates the trajectory of a single structure along its |
| 11 | +nth vibrational mode. The default mode is "path" which can be changed to "rock" |
| 12 | +using the '-ty' flag. Further customization options with each mode is described |
| 13 | +below. |
| 14 | + |
| 15 | +##################### |
| 16 | + "path" mode: |
| 17 | +##################### |
| 18 | + |
| 19 | +The most probable pathway between two equilibrium states of a protein can be |
| 20 | +calculated used the following command. |
| 21 | + |
| 22 | +path.py -f1 <start structure> -f2 <end structure> |
| 23 | + |
| 24 | +The two structures are PDB files. To successfully calculate the conformational |
| 25 | +change trajectories using the PATH program, the number of atoms in both |
| 26 | +the structures must be the same and so should the order of atoms in both the |
| 27 | +files. |
| 28 | + |
| 29 | +The program outputs the transition state (trans.pdb) and the trajectory (traj.pdb) |
| 30 | +connecting the two end states. Both these files are in the PDB format. It also |
| 31 | +outputs the energy (engy) of each structure along the trajectory relative to the |
| 32 | +two end states. |
| 33 | + |
| 34 | +The program also generates an output parameter file (path_log) which contains |
| 35 | +the RMSD between the two states of the protein, the difference in energy between |
| 36 | +the two minima, based on the "path" potential (described below), the time to |
| 37 | +the transition state from the starting structure (tbar left), time to the final |
| 38 | +state from the transition state (tbar right) and the energy of the transition |
| 39 | +state relative to the final state (Utrans). |
| 40 | + |
| 41 | +-n flag: |
| 42 | + |
| 43 | +The number of structures in the trajectory that the program outputs can be |
| 44 | +specified using the '-n' flag in the above command in the following way |
| 45 | + |
| 46 | +path.py -f1 <start structure> -f2 <end structure> -n 11 |
| 47 | + |
| 48 | +This commands outputs a trajectory with 11 structures, which includes the two |
| 49 | +end states as well. |
| 50 | + |
| 51 | +-ca flag: |
| 52 | + |
| 53 | +By default, the above commands performs an all atom simulation. But, it is also |
| 54 | +possible to run an all atom simulation or a C-alpha only simulation using the |
| 55 | +'-ca' flag. |
| 56 | + |
| 57 | +path.py -f1 <start structure> -f2 <end structure> -ca 1 |
| 58 | + |
| 59 | +This command performs a C-alpha only simulation. |
| 60 | + |
| 61 | +As PATH builds a Hessian matrix for both the structures which is later |
| 62 | +diagonalized for calculating the trajectory, the size of the protein determines |
| 63 | +how long it takes it calculate the trajectory. An all atom simulation of |
| 64 | +proteins which have less than 1000 atoms can be run on Desktop computers (8 GB |
| 65 | +RAM with i5 processor). For large systems, it is recommended to run PATH |
| 66 | +calculations on computers with larger memory. If only the overall dynamics of |
| 67 | +the protein is of interest then an alternative method to simulate large proteins |
| 68 | +is to run C-alpha only simulations. We have observed that the C-alpha only |
| 69 | +simulation can reproduce most of the essential dynamic information about the |
| 70 | +trajectory that an all atom simulation generates. |
| 71 | + |
| 72 | + |
| 73 | +##################### |
| 74 | + "rock" mode: |
| 75 | +##################### |
| 76 | + |
| 77 | +This mode of the program calculates the trajectory along the nth vibrational |
| 78 | +mode of a single structure. It can be run using the command |
| 79 | + |
| 80 | +path.py -f1 <structure> -ty rock |
| 81 | + |
| 82 | +Apart from the '-n' and '-ca' flags that behave in the same manner as in the |
| 83 | +"path" mode, there are several additions flags that could be used with the |
| 84 | +"rock" mode. |
| 85 | + |
| 86 | +-m flag: |
| 87 | + |
| 88 | +This flag can be used to specify the nth vibrational mode along which the |
| 89 | +rocking trajectory is calculated. By default the first vibrational mode is used. |
| 90 | + |
| 91 | +-exag flag: |
| 92 | + |
| 93 | +Often the magnitude of displacement along the nth vibrational mode is small and |
| 94 | +may require an increase in the magnitude to make the displacement more |
| 95 | +perceptible. This flag can use used to specify a factor that is then multiplied |
| 96 | +to the displacement. The default value is 10. |
| 97 | + |
| 98 | +-c flag: |
| 99 | + |
| 100 | +Constraints can be applied between CA atoms of pairs of aminoacids by inputting |
| 101 | +a constraints file using the '-c' flag. The constraints file consists three |
| 102 | +columns - the first column specifies the magnitude of the constraints relative |
| 103 | +to the regular force constants, and the other two columns specify the aminoacid |
| 104 | +pairs that are constrained. For example the constraint files can consist of the |
| 105 | +following lines |
| 106 | + |
| 107 | +10 A37 A43 |
| 108 | +100 A57 B74 |
| 109 | + |
| 110 | +The first line constrains the CA atoms of aminoacids 37 and 43 in chain A with a |
| 111 | +force constant of magnitude 10 times greater than the regular force constants. |
| 112 | +The second line constraints CA atoms of aminoacid 57 in chain A to aminoacid 74 |
| 113 | +in chain B with a force constant 100 times greater than the regular force |
| 114 | +constants. |
| 115 | + |
| 116 | +##################### |
| 117 | + "path" potential: |
| 118 | +##################### |
| 119 | + |
| 120 | +For all atom simulations, the program uses the ANM potential energy function |
| 121 | +described in the Chandrasekaran et al. paper (doi: 10.1063/1.4941599). For CA |
| 122 | +only simulations, the program uses a mass weighted version of the empirical |
| 123 | +potential described in the Hinsen et al. paper (doi: 10.1016/S0301-0104(00)00222-6) |
0 commit comments