http://manual.gromacs.org/documentation/5.1/onlinehelp/selections.html

Dear Himadri,

Thank you, I was already aware of this manual and this is how I came up with the command I wrote previously. Obviously, I haven't understood the manual properly so I would really appreciate if someone could indicate why the following is wrong:

gmx_d select -f *.trr -s *.tpr -select resname DMPC within 5 of resname LIG -oi select.dat

Thank you very much in advnace!

I have written the water molecule within particular distance of protein. I have used VMD for this task. You need to write a script for this purpose.

Thank you!

I know this because I've messed it up before. "gmx select" is done in nanometers. So, if you want it to be within 5 Angstroms you're going to have to do gmx select again with "0.5" instead of "5".

Also, if you're trying to get the vector of N and P atoms of DMPC, you could also call out those specifically in gmx select. You'd probably do a "resname DMPC and name P" group or something. Then use that group to say "group __ and within 0.5 of resname LIG"

Just warning you though, in the end you'll get a huge index of every every DMPC phosphate that is within 5 angstroms of your ligand for every frame, which is kind of a mess, but that's what gmx select will give.

Thank you, your answer is so informative! I am just trying to somehow measure if my drug molecule affects the DMPC molecules' conformation, if you have any better idea than my it's really welcomed! :)

Well, it kind of depends how long those DMPC lipids interact with your drug molecule. I'm just going to assume for this scenario that the interaction with any individual lipid seems to last 100 ns.

Using gmx make_ndx and the index file you'd gmx select using the "-on flag" to select indices from every ~50 ns or something. So, you'd make one index group with all of the lipids within 5 angstroms for every frame you want with a 50 ns jump. If lipids 13 and 46 were present at 50 ns and lipids 46 and 72 were present at 100 ns, then your combined group would be 13, 46, and 72.

You can then use that index group for calculating the angle in the headgroups like you planned, and also do a bunch of other things like measuring the order parameters (gmx order), diffusion coefficients (gmx msd, only if you have enough sampling though), effects on different leaflets, hydrogen bonding partners, etc.

Thanks a lot!

I am trying to do the same thing. But i am facing problem in the selection criteria.

I am trying to calculate the angle between P and N vector of my lipid (popc+popg) with regards to the z axis to see how the protein is affecting the membrane. I am using this command, but it does not working

gmx select -f *.xtc -s *.tpr -select POPG & name P* N* within 1.0 of Protein -oi number.dat

but the problem is, below command is running only 

  gmx select -f *.xtc -s *.tpr -select POPG