Klauda Lab Simulation Wiki

User Tools

Site Tools

Trace:
Writing /users/jbklauda/pub/wiki/data/cache/6/6193a558dc90ed4c190004bf67ecff70.captchaip failed
pmf_of_chemicals_1-d

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
pmf_of_chemicals_1-d [2018/01/31 15:11] – [WHAM for PMF] editpmf_of_chemicals_1-d [2018/06/29 14:49] (current) – edit
Line 1: Line 1:
 [[bilayer simulations]] [[bilayer simulations]]
  
-//under construction//+
 ======PMF of Permeation for Small Molecules====== ======PMF of Permeation for Small Molecules======
  
 Author: Eric Wang Author: Eric Wang
  
-Please email me at //ericzwang@gmail.com// if something isn't clear, unless I've graduated. +Please email me at //ericzwang@gmail.com// if something isn't clear
  
  
Line 18: Line 18:
 4. Run restrained MD\\ 4. Run restrained MD\\
 5. Get PMF with WHAM\\ 5. Get PMF with WHAM\\
-6. Get diffusion if you need 
  
  
-At this point, I will assume you have equilibrated the system in MD, the instructions for which can be found in this wiki. + 
 +At this point, I will assume **you have equilibrated the system in MD**, the instructions for which can be found in this wiki. 
  
 Before beginning, you need to plan out how you will restrain the molecules in each window. Umbrella sampling is expensive, so we insert multiple molecules into a system and hold them in different planes. This way, we can obtain multiple samples from one replicate. Look at the following excel sheet for an example Before beginning, you need to plan out how you will restrain the molecules in each window. Umbrella sampling is expensive, so we insert multiple molecules into a system and hold them in different planes. This way, we can obtain multiple samples from one replicate. Look at the following excel sheet for an example
Line 28: Line 28:
  
 Here, I have 4 ETOH molecules. //ccycle// is the window number, which I named after looking through some CGUI scripts. //zmax// is the maximum distance I sample at. This will vary depending on the thickness of your system. However, you need to have points in the water because these become reference values to 0 the PMF. You may notice that there are no negative values because I assume symmetry between leaflets. Make sure to properly space apart the molecules in each window. Big molecules will need more space, so talk to Dr. Klauda if necessary. Here, I have 4 ETOH molecules. //ccycle// is the window number, which I named after looking through some CGUI scripts. //zmax// is the maximum distance I sample at. This will vary depending on the thickness of your system. However, you need to have points in the water because these become reference values to 0 the PMF. You may notice that there are no negative values because I assume symmetry between leaflets. Make sure to properly space apart the molecules in each window. Big molecules will need more space, so talk to Dr. Klauda if necessary.
 +
 +Refer to the following path in DT1 as a reference: ''/lustre/ewang125/sc/part1/tri-cer16-32-umb''
  
 =====Building Systems===== =====Building Systems=====
Line 55: Line 57:
 Look at a few trajectories and verify that molecules have moved into proper planes. Then copy win* directories into umb directory. Remove files in win* directories except for .rea file.  Look at a few trajectories and verify that molecules have moved into proper planes. Then copy win* directories into umb directory. Remove files in win* directories except for .rea file. 
  
-Next we run the actual umbrella sampling\\+Next we run the actual umbrella sampling in the **umb** directory\\
 1. Adjust **umb.inp** similarly to how umbstrt.inp was adjusted. Talk to Dr. Klauda about what force constant is needed. I used 0.75 because 1.5 is in the literature, and CHARMM doesn't multiply by 1/2.\\ 1. Adjust **umb.inp** similarly to how umbstrt.inp was adjusted. Talk to Dr. Klauda about what force constant is needed. I used 0.75 because 1.5 is in the literature, and CHARMM doesn't multiply by 1/2.\\
 2. Adjust final run in **sdyn.csh** in the line ''$nxt > 10''.\\ 2. Adjust final run in **sdyn.csh** in the line ''$nxt > 10''.\\
Line 67: Line 69:
 =====WHAM for PMF===== =====WHAM for PMF=====
  
-Now use WHAM to extract a PMF. I use the code written at [[http://membrane.urmc.rochester.edu/sites/default/files/wham/doc.html]]. This method takes in a time series and uses it to calculate the PMF. \\+Now use WHAM to extract a PMF. I use the code written at [[http://membrane.urmc.rochester.edu/sites/default/files/wham/doc.html]]. This method takes in a position time series and uses it to calculate the PMF. \\
 Why don't I use WHAM written in CHARMM? Because there's very little documentation. The documentation in the above link is good and you could figure out how to use it yourself. Why don't I use WHAM written in CHARMM? Because there's very little documentation. The documentation in the above link is good and you could figure out how to use it yourself.
  
 Each molecule gets its own directory. For example, ETOH1 is in the e1 directory Each molecule gets its own directory. For example, ETOH1 is in the e1 directory
  
 +Position time series \\ 
 1. Adjust **pos*.inp**. Adjust path to psf,crd,dcd files. Adjust molecule segid if necessary. Increase mxa,mxs,mxt if necessary.\\ 1. Adjust **pos*.inp**. Adjust path to psf,crd,dcd files. Adjust molecule segid if necessary. Increase mxa,mxs,mxt if necessary.\\
 2. Adjust **pos*.csh**. Adjust run numbers and window range. \\ 2. Adjust **pos*.csh**. Adjust run numbers and window range. \\
-3. Submit by typing ''./pos.scr'' +3. Submit by typing ''./pos.scr''\\
 The pos*.dat files are the position time series used to calculate PMF.  The pos*.dat files are the position time series used to calculate PMF. 
  
 +WHAM \\
 1. To do WHAM, we need a meta file. It's called **meta**. Adjust the path name to pos*.dat. Adjust the z position. Adjust the force constant (multiply by 2 because this code uses 1/2 factor). \\ 1. To do WHAM, we need a meta file. It's called **meta**. Adjust the path name to pos*.dat. Adjust the z position. Adjust the force constant (multiply by 2 because this code uses 1/2 factor). \\
 2. Adjust **wham.scr**. First number is minimum window (0 in mine). Second is maximum window (40 in mine). Third is # window (41). Fourth is tolerance (use .0001). Fifth is temperature. Sixth should be 0. \\ 2. Adjust **wham.scr**. First number is minimum window (0 in mine). Second is maximum window (40 in mine). Third is # window (41). Fourth is tolerance (use .0001). Fifth is temperature. Sixth should be 0. \\
-3. Execute by typing ''wham.scr''. Output is in **pmf** \\+3. Execute by typing ''./wham.scr''. It will take only a few seconds. Output is in **pmf** \\
 4. Repeat in each directory 4. Repeat in each directory
  
 Excel \\ Excel \\
-1. The bottom section of the pmf file is the energy against window number. Using Excel, reorder the energies based on z position. \\ +1. The bottom section of the pmf file is the energy against window number. Take this section. Using Excel, reorder the energies based on z position.\\ 
-2. Zero your PMF using the end values which represent the bulk. \\ +2. Zero your PMF using the bulk water values. \\ 
-3. Calculate average and standard error using replicates. If the standard error is too large, you may need to run another replicate.   +3. Calculate average and standard error using replicates. If the standard error is too large, you may need to run another replicate. Make sure your PMF makes intuitive sense. Polar molecules should have high energy in the hydrophobic region, for example. 
-=====Estimation of Diffusion Coefficient=====+
pmf_of_chemicals_1-d.1517429467.txt.gz · Last modified: 2018/01/31 15:11 by edit