Skip to content

Analysis: alanine dipeptide

Two notebooks over the runs the cMD and REST2 pages produced.

notebook what it shows
pmf-1d.ipynb the free energy along phi, cMD against the REST2 cold rung
pmf-2d.ipynb the Ramachandran surface the 1D projection averages away

They are committed with their outputs, and rendered here as pages: every figure and number below comes from the runs these tutorials describe, not from a build-time re-execution. To run them yourself:

micromamba activate analysis-env          # see Installing, step 5
export MD_TUTORIAL_RUNS=/path/to/your/tutorial/runs
jupyter lab docs/tutorial/ALA/analysis/

The root is not defaulted, deliberately: a default would be one machine's path. A notebook that cannot find it says so by name, with the layout it expects.

What a PMF here is, and is not

F(x) = -kT ln p(x) over a sampled histogram. A free energy along ONE coordinate with everything else averaged over — not an energy barrier, and not a rate.

Empty bins stay empty. A bin nothing visited has no estimate, and filling it in would draw a barrier whose height is set by the sample size rather than by the physics. They are left as NaN.

Neither notebook claims convergence. Showing that a PMF has converged needs repeats or a block analysis, which is a different question from how to compute one.