# Constraining terminal basepairs in DNA **URL:** https://gromacs.bioexcel.eu/t/constraining-terminal-basepairs-in-dna/2549 **Category:** User discussions **Created:** [July 20, 2021, 5:05pm UTC](https://gromacs.bioexcel.eu/t/constraining-terminal-basepairs-in-dna/2549 "2021-07-20T17:05:09Z") **Posts on this page:** 1 **Showing post:** 9
### Author: ![kandresen](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/kandresen/32/747_2.png) [@kandresen](https://gromacs.bioexcel.eu/u/kandresen) #### Post date: [August 19, 2021, 4:01pm UTC](https://gromacs.bioexcel.eu/t/constraining-terminal-basepairs-in-dna/2549/9 "2021-08-19T16:01:47Z")
I just wanted to give a final update on this. I never did get the “pull” function to work. As @genie mentioned, it may have been due to my Gromacs install, but as I am running on someone else’s cluster, I was not able to modify that. I tried multiple different definitions for my groups and nothing seemed to work. However, I recently found out about the “merge” function in pdb2gmx. I used this function to combine my two DNA strands into one topology. (Previously, they were two separate topologies that were included as .itp files in the top-level topology file.) By merging the two DNA strands into one topology, I was able to create a bonded (restraint) interaction between the first and last basepairs (using the bond “10” function). Despite the fact that this should be identical (at least as far as I understand the documentation), this approach worked perfectly where the “pull” function had failed. The DNA did not melt and I did not receive any errors in the simulation. Just in case anyone is trying to do the same thing, this is the code I added to the bottom of the [bonds] section of my topology (topol.top): ```auto ; atom_i atom_j func r0 r1 r2 k(kJ mol^-1 nm^-2) 22 618 10 0.0 0.5 0.6 10000 16 615 10 0.0 0.5 0.6 10000 300 331 10 0.0 0.5 0.6 10000 303 337 10 0.0 0.5 0.6 10000 ``` The atoms indicated are the heavy atoms (nitrogen and oxygens) that form the hydrogen bonds in my terminal A-T bases. Hope this helps somebody! --- _[View the full topic](https://gromacs.bioexcel.eu/t/constraining-terminal-basepairs-in-dna/2549)._