Question about anisotropy and flexibility

Hi, if one part of a membrane protein has weaker density, and indeed the 2D plot from the angular distribution looks like there is one portion that is not particularly well covered, does this necessarily come from lack of the relative projections (I do have some preferred orientations), or this might also come from flexibility in that region ? Local refinement with mask covering that poorly defined region did not help. I ask as I recently saw the Nature paper from @olibclarke and colleagues (great stuff by the way), and anisotropy was related to flexibility, am I correct ?

Many thanks for your help !