Hi,

I would suggest making two groups for chain A and chain B in a single index file and then follow Dr. Rashmikumari's g_mmpbsa tutorial.

I guess you created two different index files because the residue numbers are not continuous. But that issue can be overcome by using atoms instead of residues while making index files like using [a 1-2000] for chain A, [a 2001-4000] for chain B.

Keerthic Aswin

Yes. I have separated based on atom numbers and I am following Dr. Rashmikumari's g_mmpbsa tutorial. I have mentioned these already in my query. You have repeated the same as the answer. Why so?

My query is, "I have created two separate ndx files with chain A and chain B separately. Now how to proceed with MMPBSA analysis?" because you can't use use two index files after -n and you can't use gmx trjcat to combine index files as we do for xtc or trr.

Dr Thirumal Kumar D

Hi, I was suggesting instead of creating two separate index files, Create two groups on the same index file. use it with -n flag and when the groups are prompted select the groups you created. I do my analysis this way and it works for me. Hope it works for you.

@keerthic

Can you please a example command on creating two groups on same index file?

Because say, i want to study number of hydrogen bonds between protein and ligand and then i proceed with options group 1 for protein and group 13 for ligand (example). So here if i create a single group of both A and B, i am not still clear with your suggestion.

Please provide command on index creation with two groups if possible

Hi Thirumal, am not sure. But used gromacs way back

I hope you can add multiple index to a same index file... Hopefully, the command will be like

make_ndx -f *.tpr

1|&r 1-20 (enter)

1|&r 21-30 (enter)

q (enter)

Once u pass the desired command with index file generated, it should prompt both the index you want to analyse

Dr Thirumal Kumar D , sure. Please try this out.

1. Assuming you already have index.ndx , Run gmx make_ndx -f md.tpr -n index.ndx -o index.ndx. First all the existing groups will be shown. To go with your example, you have 13 groups with 13 being ligand and 1 being protein.

2. Next, > a 1-2000 and enter. This will form a group called a_1-2000 as a 14th group. please enable verbose to keep track. ( Here all the atoms numbered from 1 to 2000 will be grouped but it may also include sol molecules. If so, next step, > 14 & 1. This will group all the atoms numbered from 1 to 2000 present only in group 1 (protein) into a_1-2000&protein (15th group). Rename group 15 to chain A).

3. Similar way, you can create for chain B (group 17).

4. Now you have separate groups for protein,ligand, different chains in the protein in the same index file. This index file can be parsed with -n flag with all gromacs modules and desired groups can be selected (eg. group 1 & 13 for hbond calculation and group 15 & 17 for g_mmpbsa ).

I hope this works.