Open ohsOllila opened 2 years ago
We can automatically detect the normal of the membrane. To do so we need a code that automatically separates the leaflets (this can be done by running the neighbor search for head groups only) and then calculating the moment of inertia for leaflets. Order parameter and form factor calculation can then be generalized and take the membrane normal vector as an input. If you think this is worth doing I can make a draft
This is now handled with a 'WARNING' key in the README.yaml: WARNINGS: ORIENTATION: x-normal For example, see https://github.com/NMRLipids/Databank/tree/main/Data/Simulations/a7e/c6d/a7ec6ded506e2f92f704630d74da8d5d0bf9441c/23177d5f6629268724e6e25a19047c3d3f65dc8d Systems with this key are then skipped in analyses. There are only three systems like this: a7e/c6d/a7ec6ded506e2f92f704630d74da8d5d0bf9441c/23177d5f6629268724e6e25a19047c3d3f65dc8d/ b68/566/b68566a3b3df21c923162796ed1530e2cc43427d/57c949260b6f261039d6b2839bd2e04c68fda617/ f83/806/f83806c3a0af6edc5b32cd7769ac7e689b4c5331/f0ccd45496dd8a2bb993c75e728bac817cf3bae8/
Because this a very small fraction of the total number of trajectories (currently 714), it is probably not worth of using time for this right now.
There is two simulations having membrane normal in other than z-direction. The current order parameters and form factor codes do not work for these systems: https://github.com/NMRLipids/Databank/tree/main/Data/Simulations/b68/566/b68566a3b3df21c923162796ed1530e2cc43427d/57c949260b6f261039d6b2839bd2e04c68fda617 https://github.com/NMRLipids/Databank/tree/main/Data/Simulations/f83/806/f83806c3a0af6edc5b32cd7769ac7e689b4c5331/f0ccd45496dd8a2bb993c75e728bac817cf3bae8