search

smog-server / OpenSMOG

OpenSMOG is a Python library for performing molecular dynamics simulations using Structure-Based Models. OpenSMOG uses OpenMM.
MIT License
11 stars 6 forks source link

Folding Pathway #76

Closed EBGU closed 1 year ago

EBGU commented 1 year ago

Dear SMOG team,

This is an awesome project! I really appreciate your work! I am recently working with protein folding pathway prediction, but I am quite new to this area. I wonder if I can find some examples to simulate protein folding processes with an SBM or a hybrid model. Thank you so much!

Best, Harold

contessoto commented 1 year ago

Hello, Thanks for your comment. We have the OpenSMOG documentation which contains tutorials starting from package installation, generation of input files and how to run a SBM simulation using different models. Here is the documentation link https://opensmog.readthedocs.io. I also recommend taking a look at SMOG server at https://smog-server.org OpenSMOG paper is also a good start point: https://onlinelibrary.wiley.com/doi/full/10.1002/pro.4209 Please let us know if you have additional questions.

EBGU commented 1 year ago

Thank you for your reply! Specifically, I am folding an ~250 aa protein(JMJD8). I have its structure and an extended structure generated by https://spin.niddk.nih.gov/bax/nmrserver/pdbutil/ext.html with its sequence. I generated .gro .top and .xml files for these two PDBs. I want to simulate the folding process so I run a simulation with following commands:

from OpenSMOG import SBM sbm_AA = SBM(name='JMJD8', time_step=0.002, collision_rate=1.0, r_cutoff=1.2, temperature=0.5) sbm_AA.setup_openmm(platform='cuda',GPUindex='default') sbm_AA.saveFolder('output_JMJD8_ext') sbm_AA_grofile = 'JMJD8_ext.AA.gro' #from extended pdb sbm_AA_topfile = 'JMJD8_ext.AA.top' #from extended pdb sbm_AA_xmlfile = 'JMJD8.AA.xml' #from target pdb sbm_AA.loadSystem(Grofile=sbm_AA_grofile, Topfile=sbm_AA_topfile, Xmlfile=sbm_AA_xmlfile) sbm_AA.createSimulation() sbm_AA.createReporters(trajectory=True, energies=True, energy_components=True, interval=105) sbm_AA.run(nsteps=108, report=True, interval=10**5)

However, my protein did not fold after 10**8 steps, and I did not see any secondary structure formed. I wonder if I have made any mistake, or I need a much longer simulation. Thank you!