Dear it is not as simple as most of the guys suggested you, the reason is that  both the crystal structures may be solved in different space groups. The best way is to dock as Anuraj suggested, and after that minimize the structure, if you have NAMD facility, i will suggest go for 10,000 steps to minimize. Secondly, if you have any homologue structure of the whole protein then use Modeller or other tools. 

as i understand your question it's very easy to combine two PDBs as one . simply open both the PDB in any text editor like emacs or kwrite , copy all the co-ordinates of both pdbs and paste in one file and save it as .pdb.

one more ways is to add two PDBs is that open both pdb in chimera and save structure as new.pdb.

after that you should need to clean or correct your PDB .

You can also open both structures in pymol, write in text line "select all" and save selected molecules as one file .pdb

Well in pymol.....open your first pdb file.see the structure .Open the second one again.and now click on sequence. you can see each residues and their position in the structure. I case if u are new to pymol.Open your pdb>select, hide everything>show in cartoon (your choice)> do same for the other pdb.You can compare the structure and their orientation.

Hope it may help you.

Cheers!!!

You can do this using chimera using join model option, check this out http://www.cgl.ucsf.edu/chimera/docs/ContributedSoftware/editing/editing.html#join

The second part of your question is the difficult bit. If you have no overlapping sections, you cannot use structural alignment to link the two parts. Perhaps you can find a homologous template that you can use to link the domains. If (and only if) you have reason to believe that the two domains dock against each other you can use protein-protein docking with the constraint that the two domains should connect. But I assume that there is a reason for not solving the entire structure by either group, and that reasond could very well be flexibility of the two domains.

This might sound scary (coming from a fellow bioinformatician): small-angle X-ray scattering on the whole protein can give a good indication of the domain organization.

Hi Anupam ,

you can do this in pymol by opening all the file one by one and after you will have to select them, then you will see both structures displayed as if they are aligned then save the structure or get a snap shot of the structure.

Dear Anupam,

The best way according to me is to copy and paste the x,y,z coordinates of all the atoms of amino acids of second pdb (after opening it with wordpad, or gedit(in linux) or in any other text editing program) to the first pdb after 400 residues and save it. As your first pdb was in pdb format, so after saving it, it will remain in pdb format. Just open it in Chimera or discovery studio and minimize it to resolve any bad contacts or clashes. I hope this helps.

Best Regards,

Hi Robbie,

The different parts of the proteins were crystallized for the different kind of study. For the protein-protein docking part I am not sure, but i performed a multiple structure homology modelling and performed Molecular Dynamics for the modeled structure. I need your suggestion regarding homology modeling. Conceptually it looks fine but I want to know where I can go wrong? Does protein-protein docking can be used for validation of the predict model?

The copy-paste option sounds good when you do it in chimera and join model option, but the problem is how you will be sure of relative orientation of these domains. For that, you need a similar protein having crystal structure available. If not, then take aa 300-500 from the end of first model and the start of second model, create homology models as well as ab-initio models. This part will give you an idea of the relative orientation of domains. For ab initio, Quark is the best, so far i know...

http://zhanglab.ccmb.med.umich.edu/QUARK/

Then, either homology modelling or simply align all these models and get the completed one via any software like Chimera, YASARA etc.

Hope this will help...

Hi Anupam,

It's not quite clear what you did. Do I correctly understand that you made a homology model of the full-length protein using the a homologous third protein? If that is true you can use it to align the two crystal structures and model the link between the two. No protein-protein docking needed.

Note that this is still a model. If the protein has two separate domains the relative orientation may be context dependent. That is it may change when the protein interacts with another protein.

Dear Anupam Singh,

Regarding to your question " Does protein-protein docking can be used for validation of the predict model? " ; the answer is no , you can't . If you looking for protein model like you have one protein with unknown structure and you seek find out it's 3D structure you can do it just get protein sequence of your target protein and use EXPASY srever ( http://www.expasy.org/tools/). Then you will go further study like drug design, in this case you can use Protein- protein docking.

Anupam singh,

try SymmDock. Here is the link.

http://bioinfo3d.cs.tau.ac.il/SymmDock/

Dear it is not as simple as most of the guys suggested you, the reason is that  both the crystal structures may be solved in different space groups. The best way is to dock as Anuraj suggested, and after that minimize the structure, if you have NAMD facility, i will suggest go for 10,000 steps to minimize. Secondly, if you have any homologue structure of the whole protein then use Modeller or other tools. 

Kindly help to combine the all dock pose to into single file. ..

My problem is 

1 protein

50 different ligand molecules. 

to analyse this for interactions. currently i have to prepare separate files.

Kindly help me to analyse all ligand poses from protein and all combined dock poses.

Hello everyone,

    I want to merge two .pdb files and I tried 'cat *.pdb > merged.pdb' as Saravanan suggested but I got 'command not found' . Please help me .

Thanks

I am suggesting you to go for Pymol Tool. It was extremely simple and comparatively accurate in my case.

You can open multiple structures in the same windows. which can be finally saved to same structure using only just selecting structure.

Along with that, You can also used some Discovery Studio and other commercial softwares available.

For such purposes, I use EasyModeller. I apply source fragments as templates. Aligning multiple templates, I manually adjust the alignment (taking into account the number of columns), and build complete model by fragments, as a template...

Pymol is an easy and good option for your query.

You can use Chimera, open both PDB, select the option of the combine on the model panel, that combines the files into unique pdb after that in the same panel use te option "write PDB" and use the second form of writing the pdb.

You can use VMD or pymol to align the structures and save them as one pdb. Then you need to use a software of your choice to do a minimization.