Closed primaryodors closed 6 months ago
All tests succeeding so far.
There is a lot to this PR. However, aside from a few minor fixes, all of it is interrelated and pertinent to the application of cryo-EM models to the prediction system.
All tests successful.
All functionality in this PR has either been merged, is part of an open PR, or was causing unit test failures.
When running a prediction, if any ligand-receptor clashes are observed in the active state, the directions and magnitudes (defined as ∑r(vdW) - r) of the clashes are recorded and a new temporary model PDB is generated based on adjusting the positions of the TM helices to resolve the clashes, within the constraint that the total adjustments do not exceed a fixed number of standard deviations (currently 2Σ) according to the positional variability from the cryo-EM models.