Bug report: the oxygen coordinates of peptide bond predicted by PepFlow is incorrect.

[yangxiufengsia]

Hi, thanks for such interesting paper. I played around your method, but found the generated backbone oxygen coordinate is not correct, could you please tell me the reason or help fix this bug? Thank you again.

[Guo-Stone]

To my way of thinking, this may arribute to the wrong standard position of O atom. In the dict rigid_group_heavy_atom_positions, the standard position of O is ['O', 3, (0.627, 1.062, 0.000)]. But I think it should be ['O', 3, (0.627, -1.062, 0.000)], because I have calculated it into detail.

[panEuler]

In deepmind's opensourecodehttps://github.com/google-deepmind/alphafold/blob/main/alphafold/common/residue_constants.py,the standard position of O is ['O', 3, (0.627, 1.062, 0.000)]