yuemin liu wrote: > Is that possible to model HYP(Hydroxylproline) > containing protein with PRO? I found HYP in the > modeller lib, but no one-letter code for HYP.

You would first need to define the topology and parameters for the HYP residue type. This is not straightforward. See http://salilab.org/modeller/FAQ.html#9

> Another question is that how feasible is we use > alignment from other program like blast when do > modeling.

Yes, you can use an alignment from another program as input to Modeller. You may have to edit it by hand a little first though, for example to add ligand information (most alignment methods only deal with the 20 standard amino acids) or to add the PIR header so that Modeller knows how to map your sequence to PDB structures.

Ben Webb, Modeller Caretaker