Sep 08, 2020
Minimization of protein structures using refineD
This protocol is a draft, published without a DOI.
- 1James Madison University
Protocol Citation: Chris Berndsen 2020. Minimization of protein structures using refineD. protocols.io https://protocols.io/view/minimization-of-protein-structures-using-refined-bkqfkvtn
Manuscript citation:
Bhattacharya, D. (2019) refineD: improved protein structure refinement using machine learning based restrained relaxation. Bioinformatics 35, 3320–3328.
License: This is an open access protocol distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Protocol status: In development
We are still developing and optimizing this protocol
Created: September 02, 2020
Last Modified: September 08, 2020
Protocol Integer ID: 41447
Abstract
Sometimes structural issues can be corrected with energy minimization. In this process, the weak interactions are optimized in order to produce a more thermodynamically stable structure. RefineD uses conformational sampling to try to alter homology models or other protein models to form more native like structures. This method can fix small/minor issues in the structure but cannot fix bad or inaccurate starting models.
Uploading structure
Uploading structure
Upload your structure to the interface.
Select “conservative” for refinement mode and provide your email. Then press submit.
Note
The server takes 24 to 72 hours to run typically. I recommend copying the web address on the run page to make sure you have the information in case the email gets blocked.
Analyzing refined structures
Analyzing refined structures
When the results come back, navigate to the page and review the resulting structures.
Note
There will be 1 to 5 structures returned each with two statistics. Both statistics indicate the relative change in the refined structure to the structure you submitted. Ideally they will be similar (within 5%), however it is okay if they are not.
Record the statistics and download the structures.
| Structure | GDT-HA | CA RMSD | |
| 1 | |||
| 2 | |||
| 3 | |||
| 4 | |||
| 5 |
Name your downloaded structures using the following format:
refineD_[#]_[Group_name]_[sequence_name].pdb
Replace [Group_name] with your name/group name without the brackets. Replace [sequence_name] with the name of the sequence. Replace the [#] with the refineD structure number.
Indicate your OSF file location as a link within a note on this step.
In YASARA, align your refined structures to your starting structure (the model you uploaded in step 2).
Note
It helps to color each structure differently to tell the structures apart.
After alignment, look for how refineD changed the structure by minimization and record areas that changed and if possible describe how minimization changed the structure. Look at side chain packing and position of loops/secondary structure and how these features pack.
Submit one of your refined structures to Molprobity and assess the structure.
Note
Compare the results before and after refinement to see if some of the problems identified by Molprobity were fixed by energy minimization.