Not only that, it would be even nicer to be able to thread fragments from multiple templates at the same time. (I understand that the ability to thread from templates is basically homology modeling-- but aren't fragment insertions commonly done as well?)

And while we are at it... part of the problem with threading is that, if you leave a large gap in the target sequence, sometimes you could end up with a successful threading with a really bad backbone score in some locations rather than a failed threading attempt. When this happens, local wiggling those segments with a backbone score in the negative tens of thousands would have to be done first, otherwise the displayed score would never go above zero... (For your information, viewing in full-atom mode revealed that overstretched bonds is the cause.)

These are great suggestions. We are already planning on adding the ability to thread only part of the template and to thread multiple templates. It will likely be part of a future release.

partial threading is now possible

Marking as resolved.