search

choderalab / MSMs

Markov state models
GNU General Public License v2.0
3 stars 7 forks source link

Plotting other structures onto MSM #5

Open sonyahanson opened 9 years ago

sonyahanson commented 9 years ago

I guess we need to check this a bit more thoroughly in general re: ensembler models, but currently have a basic question:

If I plot the first frame of run0-clone0.h5 is that equivalent to plotting the starting structure?

kyleabeauchamp commented 9 years ago

It's not exactly the same, but it should be close. This frame corresponds to the coordinates after equilibration, which involves some degree of structural relaxation.

sonyahanson commented 9 years ago

Cool. So I did some exploration here, and there does seem to be some discrepancy. Hard to tell if there is some dynamics that's happening super fast or if it's because we're doing something wrong. Result fir Abl using first frame of run0-clone0 vs implicit refined 2HZ0 (starting structure based on RUN0/template.txt): fig_abl_2hz0

sonyahanson commented 9 years ago

Similar for Src: fig_src_1fmk

kyleabeauchamp commented 9 years ago

Well, both points seem to lie inside a basin of high density, although the src is strange that the the points are well-separated on the slowest eigenfunction.

On Mon, May 18, 2015 at 9:38 PM, sonyahanson notifications@github.com wrote:

Similar for Src: [image: fig_src_1fmk] https://cloud.githubusercontent.com/assets/6333823/7693866/2f7bd11c-fda6-11e4-81a4-b51dda2bf4eb.png

— Reply to this email directly or view it on GitHub https://github.com/choderalab/MSMs/issues/5#issuecomment-103294760.

jchodera commented 9 years ago

Would be cool if the data munging code could include the starting structures as the first structure in each trajectory.

sonyahanson commented 9 years ago

After talking to Kyle it does seem like there was a ~10 ns equilibration before starting the simulation, which probably explains this difference.