project1

RMSD over time

Ptraj is the component of AMBER which allows you to process your trajectory, or the output coordinate file that describes the movement of your molecule over the course of the simulation. Here we will write two ptraj scripts that take in trajectories, strip out water molecules, and calculate RMSDs. We will graph the RMSDs over time to show deviation from the starting structure over the course of the simulation.

RMSD of system MD (ptraj_rmsd_all.in)

	On remote server, travel to your directory, and make a new folder named hm_analysis.
	In a terminal window, use the vi editor to open a new file called ptraj_rmsd_all.in.
	Copy the following text and paste it into the ptraj_rmsd_all.in file. Edit the solvated_model.pdb file in red to the correct name for your solvated model pdb got from the tleap. Edit the residue range in red to your correct protein residue number range. Save this file to your hm_analysis directory. # Calcualte RMSD from the initial frame (all) trajin ../md_simulations/all_prod.crd reference ../md_simulations/solvated_model.pdb strip :WAT rms reference out ptraj_rmsd_all.out :1-132 go
	Navigate to your hm_analysis directory. At the command line, enter the following text, with solvated_model.top in red changed to the correct name for your solvated model top file get from the tleap program. $ cpptraj ../md_simulations/solvated_model.top <ptraj_rmsd_all.in> ptraj_error_all.out
	Copy ptraj_rmsd_all.out to your local computer(Mac) and open this file in Microsoft Excel as delimited by spaces and graph Overall RMSD vs. Time.

Temperature over time

Here we will extract and graph data from our MD simulation outputs to observe the fluctuation of system temperature over the course of the simulation.

In a terminal window, navigate to your hm_analysis directory and enter the following text.

$ grep TIME ../md_simulations/md_all.out > temp_time.txt

Open the temp_time.txt file in Microsoft Excel as delimited by spaces and graph Temperature vs. Time.

Total energy over time

Here we will extract and graph data from our MD simulation outputs to observe the fluctuation of total system energy over the course of the simulation.

In a terminal window, navigate to your hm_analysis directory and enter the following text.

$ grep Etot ../md_simulations/md_all.out > etot.txt

Open the etot.txt file in Microsoft Excel as delimited by spaces and graph Total Energy vs. Time. (Get time from temp_time.txt.)

Trajectory movie

The trajectory from your MD simulation is really just a series of sequentially related snapshots and can therefore be thought of as a movie. We will use VMD to render our trajectories in an actual movie format.

At first , you need to copy your solvated *.prmtop and all_prod.crd file to you local computer with scp command.

In Mac, search VMD in Lauchpad and double click VMD to open it.

	Go to File->New Molecule. Click Browse and select the top file of your solvated homology model from your md_simulations directory. Select AMBER7 Parm as the file type in the Determine file type: dropdown menu and click Load. (You will not see a structure load into the VMD window yet.)
	Keep the top file in the Load files for: box instead of choosing new molecule. Click Browse and select the all_prod.crd file from your md_simulations directory. Select AMBER Coordinates with Periodic Box as the file type in the Determine file type: dropdown menu and click Load. (Now you should see your structure loaded into the VMD window. It should be moving because you are displaying the trajectory of your MD simulation.)
	While the trajectory loads, make a pretty representation of your model to be used in the trajectory movie. Click Display->Orthographic and Display->Axes Off. Go to Graphics->Representations and replace all in the selection box with all and not water and hit enter. This should hide the water molecules. Make a Cartoon or New Cartoon colored by structure representation of the model. Rotate your model around and make observations. Find a way to orient the model that allows you to see the interesting features of the simulation.
	Once the trajectory has finished loading and you have a representation/orientation of the model that you like, go to Extensions->Visualization->Movie Maker. In the VMD Movie Generator, set the parameters to the following: Renderer: Snapshot Movie Settings: Trajectory Format: MPEG-1 Working Directory: Your hm_analysis directory Name: Give the movie a name Rotation angle: Leave as default Trajectory Step Size: 5 Duration: Leave as default Click Make Movie to begin making a movie of your trajectory. The process should take about 5 minutes. You are expected to use this movie in your Project 1 presentation.

The refined homology model

Unlike your initial homology model, which is represented by a convenient single structure, your refined homology model is described by a trajectory, or series of sequentially related structures. In order to obtain a single structure and data set that can be compared directly with the initial model, we will calculate averages from the refined model trajectory.

Calculate the average structure

	We focus on the protein ,not water ,so here we need to get a no solvated model with tleap program in md_simulations folder. $ vi get_no_wat.in Copy the following text and paste it into the get_no_wat.in file. Save this file to your hm_analysis directory. source oldff/leaprc.ff99SB aaa = loadpdb final.pdb saveamberparm aaa final_no_wat.prmtop final_no_wat.inpcrd quit And run tleap program with get_no_wat.in as input script. $ tleap -f get_no_wat.in
	And for the trajectory of MD, we also need to remove water. At here , we can use cpptraj to do it.
	First, we will use cpptraj to calculate the average structure based on all the different structures observed over the course of the simulation. Note that the average structure is not real! Travel to hm_analysis folder ,In a terminal window, use the vi editor to open a new file called ptraj_get_avg_struct.in. $ vi ptraj_get_avg_struct.in
	Paste the following text into the ptraj_get_avg_struct.in file and edit the residue range in red to the correct residue range for your protein. Save the file to your hm_analysis directory. # Get average structure trajin ./all_prod_nowat.crd average traj_avg_struct.pdb :1-132 start 50 stop 3050 offset 1 pdb go
	Run the ptraj script by entering the following text at the command line. Edit the no_solvated_model.top file in red to the correct filename of your no_solvated_model.top. $ cpptraj ../md_simulations/no_solvated_model.top < ptraj_get_avg_struct.in > error_get_avg_struct.out

Find the structure most like the average

Next, we will calculate RMSD from the average structure over the course of the simulation. From this, we can see which structure from the MD is the best approximation of the average structure. This way we get an "average" structure that is real.

	In a terminal window, use the vi editor to open a new file called ptraj_rmsd_from_avg.in. $ vi ptraj_rmsd_from_avg.in
	Paste the following text into the ptraj_rmsd_from_avg.in file and edit the residue range in red to the correct residue range for your protein. Save the file to your hm_analysis directory. # Calculate RMSD from average structure trajin ./all_prod_nowat.crd reference ./traj_avg_struct.pdb rms reference out ptraj_rmsd_from_avg.out :1-132 go
	Run the ptraj script by entering the following text at the command line. Edit the final_no_wat.prmtop file to the correct filename of your no water .top file model. $ cpptraj ../md_simulations/final_no_wat.prmtop <ptraj_rmsd_from_avg.in> error_rmsd_from_avg.out*
	Copy the ptraj_rmsd_from_avg.out file to your local computer and open it in Microsoft Excel as delimited by spaces and graph RMSD vs. Time. Use the graph to identify the time step of the structure with the lowest RMSD from the average structure. This will be the final structure you will use to represent your refined homology model.

Obtain snapshot of the structure most like the average

Finally, we will extract the structure we will use to represent our refined homology model from the 3000 others in the trajectory.

In a terminal window, use the vi editor to open a new file called ptraj_get_struct_like_avg.in.

Paste the following text into the ptraj_get_struct_like_avg.in file and edit the numbers in red to the correct time step of the structure with the lowest RMSD from the average structure. Save the file to your hm_analysis directory.

#Get snapshot most like the average structure
trajin ./all_prod_nowat.crd # # 1
trajout struct_like_avg.pdb pdb
go

Run the ptraj script by entering the following text at the command line. Edit the final_no_wat.prmtop file to the correct filename of your no water *.top file model.

$ cpptraj ../md_simulations/final_no_wat.prmtop < ptraj_get_struct_like_avg.in > error_get_struct_like_avg.out

Evaluate the rationality of the model

	We will choose the structure most like the average of the simulation to submit to the Mol Probity website to obtain Ramachandran plots of our models. Copy this file to your local computer. Go to the website and upload the file (file name) we just copied from server. Once the file has been uploaded, you will see “the job is running”. Click Continue.
	In Main page, click Analysis all-atom contacts and geometry(ensemble).
	The Analyze all-atom contacts and geometry (ensemble) window will be displayed. Go to the page footer, leave the defaults selected , and click Run programs to perform these analysis.
	On the Analysis output page, click View next to Ramachandran plot PDF. Save this PDF to your hm_analysis directory.