Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Possibly incorrect Jacobian in RMSD using atomPermutation #871

New issue
New issue
Open
Open
Possibly incorrect Jacobian in RMSD using atomPermutation#871
@HanatoK

Description

@HanatoK
HanatoK
opened on Oct 22, 2025

In my opinion, the code in

colvars/src/colvarcomp_distances.cpp

Lines 1075 to 1082 in 2e87649

cvm::atom_pos &y = ref_pos[ia];
for (size_t alpha = 0; alpha < 3; alpha++) {
for (size_t beta = 0; beta < 3; beta++) {
rotation_term += grad_rot_mat[beta][alpha][alpha] * y[beta];
// Note: equation was derived for inverse rotation (see colvars paper)
// so here the matrix is transposed
// (eq would give divergence += grad_rot_mat[alpha][beta][alpha] * y[beta];)
should select the reference positions with the best permutation. I am not sure, though.

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions