Empty structures for transmembrane protein

[peters.m...@gmail.com]

Thanks a lot for putting up the server!
I tried to get a model for an ion channel, initially only the pore domain (2 transmebrane domains and the pore itself), and then the whole transmembrane part of the channel. In both cases the server returned an empty structure (pdb file with 0 byte size). For another cytosolic part of the channel the prediction worked.
When I download the "Complete result package" there are 2x5 different pdb files (Name_70/100_1-5_hMIN). I assume those are the results with different numbers of constraints.