Minimized pdbs contain non pdb format lines at the end

[rsanchezgarc]

At the end of the file, after the connect records, I see that we have the energetics computed by rosetta. However, this part can cause problems, for instance, when using oddt to load the pdb. Wouldn't it be useful to remove or to add the #REMARK symbol?

CONECT 3200 3197 3198                                                           
# All scores below are weighted scores, not raw scores.
#BEGIN_POSE_ENERGIES_TABLE 
label fa_atr fa_rep fa_sol fa_intra_rep fa_intra_sol_xover4 lk_ball_wtd fa_elec pro_close hbond_sr_bb hbond_lr_bb hbond_bb_sc hbond_sc dslf_fa13 omega fa_dun p_aa_pp yhh_planarity r
ef rama_prepro total
weights 1 0.55 1 0.005 1 1 1 1.25 1 1 1 1 1.25 0.4 0.7 0.6 0.625 1 0.45 NA
pose -1176.44 156.416 774.501 2.33027 46.5727 -31.3098 -278.709 2.9261 -88.2726 -9.30775 -41.1764 -17.4642 0 7.466 473.713 -31.8512 0.00277 32.4346 -19.753 -197.921
MET:NtermProteinFull_1 -3.82669 0.31211 2.01105 0.01699 0.39491 -0.18555 0.44667 0 0 0 -1.11942 0 0 0.07214 3.08174 0 0 1.65735 0 2.86132
ARG_2 -1.64531 0.24283 1.26034 0.01668 0.58575 -0.0887 0.90376 0.25294 0 0 0 0 0 0.00075 5.30527 -0.06761 0 -0.09474 2.76409 9.43604
PRO_3 -4.90234 1.48984 2.83508 0.00325 0.09837 0.01641 -0.70188 0.58005 0 0 0 0 0 0.10497 0.18928 -0.99188 0 -1.64321 2.65007 -0.27198
PRO_4 -3.54852 1.50816 1.88982 0.00237 0.07537 -0.11618 -0.0056 0.24017 0 0 0 0 0 -0.07361 2.71735 -0.61055 0 -1.64321 -0.29276 0.14281
GLY_5 -2.6971 0.37045 3.01958 0.0001 0 -0.27099 -0.97261 0 0 0 0 0 0 0.278 0 -1.13567 0 0.79816 -0.54917 -1.15924
TYR_6 -8.80802 1.34854 3.52327 0.02118 0.09237 -0.57155 -0.5129 0 0 0 0 0 0 0.07983 2.57859 0.08207 0 0.58223 -0.61601 -2.2004
GLU_7 -6.14907 0.65412 4.58045 0.00389 0.19168 0.1942 -2.00072 0 0 0 0 -1.07148 0 0.0235 3.36692 0.04556 0 -2.72453 -0.22442 -3.10991
SER_192 -2.00534 0.10835 1.44188 0.00276 0.05504 -0.14393 -0.2593 0 0 0 0 0 0 -0.06956 1.14332 -0.11066 0 -0.28969 -0.03476 -0.1619
PHE:CtermProteinFull_193 -7.45183 10.5097 2.98893 0.02111 0.33799 -0.28575 -0.34418 0 0 0 0 0 0 0 2.44633 0 0 1.21829 -0.03794 9.40262
LIG_194 -12.7114 1.37937 0.56323 0.04739 0.13413 0.25608 0 0 0 0 0 0 0 0 0 0 0 0 0 -10.3312
VRT_195 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
#END_POSE_ENERGIES_TABLE 

[matteoferla]

That is very much a standard Rosetta thing. It can be controlled pose.dump_pdb('👾👾👾.pdb') as opposed to pose.dump_scored_pdb('👾👾👾.pdb'). When I deal with #29, I will add the option for the latter to be opt in.

The REMARK option would be nice and even though a PyRosetta pose does keep its header data, it is problematically implemented and can be dangerous as it is prone to segmentation faults or infinite loops. So a REMARK route would require crude text surgery, hence my preference of the scores to be opt in. Plus, the POSE_ENGERGIES_TABLE may be munged by a 3rd party downstream script (e.g. residue energy to b-factors for PyMOL etc.)