some i-rmsd values in haddock stats files are not correct

[LilySnow]

This issue is not related with deeprank code directly. It is to record a data error introduced when we use haddock to generate data.

Some i-RMSDs reported in XXX.stats do not seem to be correct. For example, 1ACB_cm-it0_9323.pdb has an irmsd of 8.888 in 1ACB.stats, but deeprank calculated 17.632. I used ProFit and got the same number as deeprank. I got the refe pdb and the model pdb from 1ACB.tgz, and i-zone files from HADDOCK-ready/1ACB/ana_scripts/target.izone I put all these files under /projects/0/deeprank/BM5/irmsd_debug on cartesius.

This error is confirmed by Alex.

[amjjbonvin]

Well - this is not a DeekRand code issue. Not doesn't make sense to open an issue for this here. It's more for us to correct those values before we make any decoys public.