You are here

Fluctuations in I_sc

5 posts / 0 new
Last post
Fluctuations in I_sc


I am using "local  refinement" to calculate the I_sc of protein complexes. When I ran Rosetta for the exact same structure multiple times and I got a significant fluctuation in I_SC score, ranging from -2 to -15. This means that in one trial, the complex is unfavorable, but it is favorable in the subsequent trial. What is the reason for this? How can I overcome it? 

Thank you in advance,


Post Situation: 
Thu, 2018-09-13 12:07


That shouldn't be happening. Are you sure that the poses that are scored are the exact same ones? Minor differences could cause 13 point fluctuations at the interface. Could you share with us the command line options that you used? That'll help in debugging.



Tue, 2018-09-18 09:36

Dear Shourya,

Thank you for your prompt response! I also didn't expect such a huge fluctuation, but this is what I observed. Attached you can find the PPI structure that gave very different results (I am using the exact same structure in multiple runs). The flags that I am using are below:

For prepack:

"%s -database %s -s %s -partners %s -ex1 -ex2aro -out:file:scorefile %s/ -overwrite -detect_disulf false"

For local refinement:

"%s -database %s -s %s -docking_local_refine -partners %s -ex1 -ex2aro -overwrite -detect_disulf false -out:path:score %s"

Thank you for your help!


File attachments: 
Thu, 2018-09-20 07:33

I am also attaching the score files and the  output structures of 2 runs as an example. 

Fri, 2018-09-21 07:41

Here is the second output structure. Although the input structure exactly the same for both runs, the outputs are very different from each other, interms of I_sc and strucutre.

Fri, 2018-09-21 07:42