This repository contains code and examples to model multiple missing loops in tetrameric protein.
Modeller script is mainly based on following link: https://salilab.org/modeller/wiki/Missing%20residues
| Original resid | Afterwards | Original resid | Afterwards |
|---|---|---|---|
| 561 | 561-74=487 | 589 | 589-74=515 |
| 416 | 416-74=342 | 427 | 427-74=353 |
| 693 | 693-74=618 | 698 | 698-74=624 |
../../bin/mod9.21 mod-seg.py to execute the python script for Modeller. mod9.21 is an executable version of Python and Modeller.
- Execute the loop modeling for each chain. Mutate the "A" letters in `two-alignment.ali` to "B"/"C"/"D" letters. And run them in separate folder, because the output PDB files have the same names. Later I visually checked each chain's modeling result and `cat` them together.
List of output files as example:
- Loop modeling results: loops forming a knot -- need to fix
Two segments have knotted structure of peptides. In order to prepare a cleaner version where there is no such problem, I use CHARMM script to solve this bad penetration issues. The program requires feeding of PSF and CRD files. So I generated PSF and CRD file firstly by CHARMM-GUI's PDBReader. Later I used following script to rotate the loops and did some minimization to avoid clash and bad angles.
Read input structures:
- step1_pdbreader.psf
- step1_pdbreader.crd
Processing CHARMM script:
- step2_pene_solve.inp
Principle:
- Set some parts movable and others fixed;
- Rotate, translate the movable part;
- Local minimization first, then global minimization;