RMSD for predictive protein structures

Question

[garimav89]

I prepared a native structure of my protein from iTasser since it did not have a crystal structure. The length of the protein is >1000. Next I inserted a single amino acid mutation in it and again modelled through iTasser. The mutated structure is completely different. When I checked the RMSD for the native with mutated structure by TM-score, I'm not able to interpret the results.

They are as follows:

The TM-score is good but I cant understand RMSD values (RMSD=72.6) which is not comparable with the superimposed structure (RMSD=1.43).

Is there any problem with my native structure or mutated?

Structure1: A217724     Length= 1088
Structure2: B217724     Length= 1088 (by which all scores are normalized)
Number of residues in common= 1088
RMSD of  the common residues=   72.639

TM-score    = 0.5050  (d0=10.89)
MaxSub-score= 0.4017  (d0= 3.50)
GDT-TS-score= 0.4278 %(d<1)=0.3658 %(d<2)=0.3952 %(d<4)=0.4504 %(d<8)=0.5000
GDT-HA-score= 0.3810 %(d<0.5)=0.3125 %(d<1)=0.3658 %(d<2)=0.3952 %(d<4)=0.4504

 -------- rotation matrix to rotate Chain-1 to Chain-2 ------
 i          t(i)         u(i,1)         u(i,2)         u(i,3)
 1   -175.6655165677   0.8438927700  -0.0706573920   0.5318388156
 2    110.5973926706  -0.5363370513  -0.0857983922   0.8396315879
 3    308.5374717331  -0.0136952629  -0.9938038885  -0.1103008201

Superposition in the TM-score: Length(d<5.0)=496  RMSD=  1.43