To generate pdbs of each clusters at a particular timescale

[Bazzinga18]

Hii, i want to generate pdbs of each microstate at a particular timescale, please help me out. I already tried few codes as i am new to python and pyemma it takes so much time for me to understand, and i am not able to generate the pdb:

clustering = pyemma.coordinates.cluster_kmeans(Y,k=15, fixed_seed=1)
dtrajs = clustering.dtrajs
its = pyemma.msm.its(dtrajs, lags=1000, nits=10, errors='bayes')
pyemma.plots.plot_implied_timescales(its, units='ms', dt=0.4)
msm = pyemma.msm.bayesian_markov_model(dtrajs, lag=20, dt_traj='0.4 ns')
eigenval = msm.eigenvalues()
L=msm.eigenvectors_left(6)
ev0=L[0,:].real
EV = ev0.tolist()
pyemma.coordinates.save_traj(EV, indexes, outfile='ev01.pdb', top=pdb)

so i want to ask what should be the indexes here if i want to generate pdbs at the slowest timeperiod i.e., at ev0? thanks

[clonker]

EV is not the right input parameter to save_traj, please see the documentation. You'll want to sample frames for each discrete states (for example with pyemma.util.discrete_trajectories.sample_indexes_by_state) and then based on these indices save trajectories.

[Bazzinga18]

Okay and what about indexes?

[clonker]

Hi, you can give it a numpy array of shape arr.shape == (1, 2) with arr[0, 0] = trajectory_index and arr[0, 1]= frame_index. This you can call in a loop with different output filenames or use save_trajs.

[stale[bot]]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.