# GROMACS SMD for TCR–pMHC: should TCR be pull group 1 or 2

**URL:** <https://gromacs.bioexcel.eu/t/gromacs-smd-for-tcr-pmhc-should-tcr-be-pull-group-1-or-2/13414>\
**Category:** User discussions\
**Tags:** mdp-parameters, simulation-setup\
**Created:** [April 25, 2026, 10:04pm UTC](https://gromacs.bioexcel.eu/t/gromacs-smd-for-tcr-pmhc-should-tcr-be-pull-group-1-or-2/13414 "2026-04-25T22:04:25Z")\
**Posts on this page:** 1\
**Page:** 1

<div class="post-metadata">

**Author:** ![rachana](https://avatars.discourse-cdn.com/v4/letter/r/4da419/32.png) [@rachana](https://gromacs.bioexcel.eu/u/rachana)\
**Post date:** [April 25, 2026, 10:04pm UTC](https://gromacs.bioexcel.eu/t/gromacs-smd-for-tcr-pmhc-should-tcr-be-pull-group-1-or-2/13414/1 "2026-04-25T22:04:25Z")

</div>

GROMACS version:  
GROMACS modification: Yes/No  
Here post your question

I am setting up constant-velocity SMD for a TCR–pMHC system. The C-terminal Cα of the MHC α-chain is position-restrained as the anchor, and the pulling coordinate is defined by the vector connecting the CoM of the TCR and the MHC α-chain, aligned along the x-axis. A virtual harmonic spring is attached to the TCR CoM and moved along +x at constant velocity.

My question is:

**In the GROMACS pull code, should the TCR be pull group 1 and the MHC be pull group 2 with** pull\_coord1\_groups = 1 2, or is it better the other way around?

This is my current pull set up:

```auto
pull = yes ; Center of mass pulling will be applied on 1 or more groups using 1 or more pull coordinates.
pull_ncoords = 1 ; we have only one reaction coordinate 
pull_ngroups = 2 ; two groups defining one reaction coordinate 
pull_group1_name = TCR 
pull_group2_name = MHC_anchor 
pull_coord1_type = umbrella ; harmonic potential
pull_coord1_geometry = direction ; Pulling along x-axis 
pull_coord1_dim = Y N N ; 
pull_coord1_vec = 1 0 0
pull_coord1_groups = 1 2
pull_coord1_start = yes ; define initial COM distance > 0
pull_coord1_rate = 0.01 ; 0.01 nm/ps = 10 nm/ns
pull_coord1_k = 100 ; kJ mol^-1 nm^-2
pull_print_components = yes ; 
pull_print_ref_value = yes
pull_nstxout = 500 ; value taken from paper
pull_nstfout = 500

```

Also, When I used

pull\_group1\_name = TCR

pull\_group2\_name = MHC\_anchor

pull\_coord1\_groups = 1 2

pull\_coord1\_vec = 1 0 0

pull\_coord1\_rate = +0.01

My smd\_pullx.xvg, the coordinate starts negative (~ -4.95 nm) and only becomes positive after ~800 ps. For example:

t = 0 ps → -4.95 nm

t = 500 ps → -2.09 nm

t = 800 ps → ~0 nm

t \> 800 ps → positive

Also, I explicitly verified the orientation beforehand using MDAnalysis, and the MHC → TCR vector is aligned along +x (TCR is physically to the right of MHC).

Does this mean that the pull coordinate behaves like:

r ≈ COM(MHC) − COM(TCR) which is opposite of what I expected from the group1/group2 definition), or is there is some other reason for it?