I think that you can check the unfolded structures first. After all, the modeller just did homology modelling. I guest your align file should like this:

>P1;proteinA structureX:proteinA aaaaaaaaaaaaaaaaaaaaaaaaaaaa----------------------------------* >P1;proteinB structureX:proteinB ----------------------------bbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb* >P1;chimera sequence:chimera aaaaaaaaaaaaaaaaaaaaaaaaaaaabbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb*

But on this topic, I have a question. If I want to get a model from tweo different protein , how can I obtain satisfactory relative orientation of modeled domains in such cases, orient the two template structures appropriately before the modeling ?

Sincerely, Shiyong MDL in Peking University Tel1:+86(0)62757520 e-mail:syliu@mdl.ipc.pku.edu.cn http://mdl.ipc.pku.edu.cn/~syliu

On Thu, 17 Jun 2004, Benoit Charloteaux wrote: > Dear Modeller users, > > I want to build a model of a chimeric protein based on two distinct > structures. I have tried as explained in the FAQ section of the manual > (question 4), but I only get completely unfolded structures. > > Is it possible to build a chimeric model with this procedure? Where does > this problem come from? > Any possible explanation or suggestion would be greatly appreciated! > > To check my procedure, I have tried the following : > > I used an older prediction that worked and which is based on a single > template structure. > > Target xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx > proteinA aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa > > > I made a copy of the template structure file and renamed it to do as if > there was two distinct template structures. I tried to build a kind of > chimeric model based on these template structures as follow : > > Target xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx > proteinA aaaaaaaaaaaaaaaaaaaaaaaaaaaa---------------------------------- > proteinA -------------------------aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa > > Since the two template structures used for the prediction correspond to > the same file, the relative orientation of both structures is correct > (the atom coordinates are the same). > > The predicted 3-D model is completely unfolded too. > > Sincerely, > > CHARLOTEAUX Benoit > Centre de Biophysique Moleculaire Numerique > Faculte Universitaire des Sciences Agronomiques de Gembloux > Passage des deportes, 2 > B-5030 Gembloux > Belgium > Tel1:+32(0)81/62.25.32 > Tel2:+32(0)81/62.25.21 > Fax:+32(0)81/62.25.22 > e-mail: charloteaux.b@fsagx.ac.be > http://www.fsagx.ac.be/bp/ > > > _______________________________________________ > modeller_usage mailing list > modeller_usage@salilab.org > http://salilab.org/mailman/listinfo/modeller_usage >