Skip to content

Accelerating protein structure refinement speed of single chain protein from AI based generative model.

Notifications You must be signed in to change notification settings

Kexin-Zhang-UCAS/alphafold2mmopt

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

2 Commits
 
 
 
 
 
 
 
 

Repository files navigation

alphafold2mmopt

Based on alphafold2, alphafoldmmopt is developed to optimize molecules from any protein-generation model.

dependence

conda install openmm
pip install dm-tree
pip install biopython
pip install dm-haiku
conda install -c omnia pdbfixer
pip install pytorch

installation and run

  • install
pip install alphafold2mmopt
  • running: for a given any *.pdb file
# shell
alphafold2mmopt *.pdb
# python
from proteinopt.common.protein import Protein
from proteinopt.relax.localminimizer import relax
prot = Protein(pdbname)
opt_pdb=relax(prot)

Updation

v0.2

  1. Use pytorch instead of Jax to accelerate lDDT matrix calculation.

The lDDT calculation time can be reduced from 3s to 0.1s.

  1. Reconstruct api for Protein object and reduce repeated clean operation.

About

Accelerating protein structure refinement speed of single chain protein from AI based generative model.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages