I am modeling a protein which has a number of PDB homologes deposited. One is very high resolution, and another is of lower resolution. For every sequence I have submitted, I-TASSER ranks higher the model created using the low resolution structure as a template than that created from the high resolution structure.
for example, in the results of one of my runs:
http://zhanglab.ccmb.med.umich.edu/I-TASSER/output/S54857/
1qh1 is the lower resolution structure from one organism, and 1m1n is a higher resolution structure from a different organism.
this happens for a wide range of sequences - thus it is likely not due to the query sequence being more closely related to the low resolution structure in sequence space.
how can this be understood?