# Gmx sasa command

**URL:** <https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924>\
**Category:** User discussions\
**Tags:** analysis-tools\
**Created:** [April 8, 2021, 7:09pm UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924 "2021-04-08T19:09:00Z")\
**Posts on this page:** 5\
**Page:** 1

<div class="post-metadata">

**Author:** ![deligkaris](https://avatars.discourse-cdn.com/v4/letter/d/dbc845/32.png) [@deligkaris](https://gromacs.bioexcel.eu/u/deligkaris)\
**Post date:** [April 8, 2021, 7:09pm UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924/1 "2021-04-08T19:09:00Z")

</div>

GROMACS version: 2018.1  
GROMACS modification: No

Hi everyone, in one of my MD simulations with a small molecule in water, gmx sasa results in the atom’s SASA to be always a multiple of 0.038 (0, 0.038, 0.075, 0.0113). The SASA of the atom always takes one of these discrete values for the entire MD simulation. I noticed the same thing for another atom of that small molecule.

Is this an indication that I am running the sasa command wrong? The options I use are -surface ‘small molecule’ -output ‘atom of small molecule’

---

<div class="post-metadata">

**Author:** ![alevilla](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/alevilla/32/439_2.png) [@alevilla](https://gromacs.bioexcel.eu/u/alevilla)\
**Post date:** [April 9, 2021, 8:24am UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924/2 "2021-04-09T08:24:16Z")

</div>

Hi,

If you want to calculate the accessible solvent surface of a molecule in solution and output the atom information you can run gmx sasa with option -ao.

I recalled that you post a similar question before. But I could not understand why you use -surface.

Best regards  
Alessandra

---

<div class="post-metadata">

**Author:** ![deligkaris](https://avatars.discourse-cdn.com/v4/letter/d/dbc845/32.png) [@deligkaris](https://gromacs.bioexcel.eu/u/deligkaris)\
**Post date:** [April 9, 2021, 7:37pm UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924/3 "2021-04-09T19:37:49Z")

</div>

Thank you Alessandra! I decided to repost my question in an abbreviated form in case my first post was not easy to understand…

The -oa option provides only the average and standard deviation. With the options I am using, I get the SASA over time which I can then analyze and get the frequency distribution (which is what I wanted to do).

I do not know why the atomic SASA takes only discrete values during my 1.5 us simulation.

The -oa option does work fine but it provides only an average…

---

<div class="post-metadata">

**Author:** ![jalemkul](https://dub1.discourse-cdn.com/flex017/user_avatar/gromacs.bioexcel.eu/jalemkul/32/18_2.png) [@jalemkul](https://gromacs.bioexcel.eu/u/jalemkul)\
**Post date:** [April 9, 2021, 7:50pm UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924/4 "2021-04-09T19:50:11Z")

</div>

There are only going to be so many grid points on the surface that can be assigned to any atom, so if you’re just computing the SASA of a single atom, it seems quite likely to only assume discrete values. You can see if it varies by increasing the value of `-ndots` when executing `gmx sasa`. If it does, you have your answer, but I think this is likely it.

---

<div class="post-metadata">

**Author:** ![deligkaris](https://avatars.discourse-cdn.com/v4/letter/d/dbc845/32.png) [@deligkaris](https://gromacs.bioexcel.eu/u/deligkaris)\
**Post date:** [April 9, 2021, 8:16pm UTC](https://gromacs.bioexcel.eu/t/gmx-sasa-command/1924/5 "2021-04-09T20:16:31Z")

</div>

That took care of it, thank you Justin.