Need more specific data of your requirement for a possible solution.

I predicted a homology model, but one of my loop is long and with a less alignment to my template, so I should remodel it without template.

Since it is long (i.e., 20 residues ), there is probability of a short helix .

I preformed using Rosetta loop modelling denovo design on the base of fragment data base and generated 1000, loop models, but I can not generated a short helix in my loop.

Thanks

If you are sure that sequence patch you are focusing on is a helix not a loop then simply extract the patch and make it a query, perform blast-p (PDB) and look for the best match. Cross check the matched structure that in which template such fragment is forming a helix. Pick the same structural part from the detected template and use it in modeller as a second template to make it as a helix insert at the same location (right alignment is important in this case).

I am sure with modeling with two template and a insert in between but BLAST part is purely hypothetical work flow on your provided information. While performing blast you can vary (down) word size to find a better match in case of no detectable similarity with defaults. Also mind the changing of other parameters as required.

Dear Rajnikant,

I found the fragment that can be substituted as a helix to my loop.

But would you please let me know how can I place it to my predicted model by modeller,

I did not use modeller yet,

or which python script we use for this remodelling.

Thanks,

Dear Ramin,

below description might help you and if not send me your files. I will try to do it for you.

Your whole seq.

AAAAAAAAAAAAAAAAALLLLLLLLAAAAAAAAAAAAAAAAAAAAAAAAAA

L is the patch you predicted to be helix. So the steps could be

1. Pick the seq patch. ALLLLLLLLA

2. BLAST against PDB. Lets assume BLAST returned PDB id A23D

3. Visualize the structure and manually inspect the aligned part is a helix or not. If yes Copy the structural information (Simply helix part).

4. Prepare Modeller Alignement file like image attached.

Here modeller read all structural info from first template except your patch that it will pick from second template. And of course in your second template the structure is helix so it will insert and model that part as a helix.

No specific python script is required other then Automodel Python script. Simply mention your template name with a " , " comma.

That's all.

I think you could make a molecular dynamics simulation on your peptide, in this way you would have a better predictive model

Thanks dear Rajnikant, I am trying to follow your kind instruction .

Dear Giosue,

Yes your are right but for now, I do not want to preform MD.

Hi Dear Rajnikant,

I am trying to locate the short helix in the loop (i.e., I got from crystal structures) via multiple comparative modelling of the modeller, but it does not superimpose my helix to the loop.

Dear Ramin, didnt get you clearly.

Dear Rajnikant,

It is the same (i.e., our last discussion)

I am trying to put a short helix into my loop, so I am employing comparative modeling of modeller.

two template :

1. my protein (i.e., includes loop).

2. short helix.

3.target.

during building, auto modeller did not super impose the two templates .

I think you didnt get the idea correctly. Review my second descriptive ans carefully. See i will try to explain in simple words. It is almost replacing your target loop part from the identified template for helix part only. See in your description you told that you have already structure of your target. Use the same structure as first template to regenerate the structure of your target except that identified part which is a loop in your target structure. Now simply delete this loop structure information from your target structure and to compliment this part use another template which contains similarity with this part and forms a helix.

Hope this will solve. If not send me your files. I will try to do this for you.

Wishes.

Yes, I did the same thing.

I attach files, please find.

Thanks so much,

Ramin

Let me work on. Revert you after 24 hrs.

Thanks Dr. Rajnikant,

I did the alignment, that I am trying to import that short helix to that place that I aligned with, that is clear in *Drd2-Drd2.ali* file,

Dear Ramin,

Can you please mention here your query sequence and within that your predicted part as helix. As per your alignment it is a little tedious for me to get the right portion of sequence.

Dear Ramin,

I tried to identify the helix part on the basis of your given files and prepared fused model (attached) but the portion is again a coil. Visual inspection revealed the 3D impossibilities and that requires loop refinement protocols. Still i am unable to understand why modeller is interpreting the 2nd template information as a coil rather a helix as given in structural information. Well let me work on this issue. If you find a suitable explanation please report at [email protected]

Dr. Rajnikant ,

I appreciate you giving your time from you busy schedule.

Yes. I was faced with the same problem,

Modeller claimed (http://salilab.org/modeller/9.12/manual/node21.html) that for comparative modeling via multiple template does not need superimposing, but as you see, it can not build it without initial superimpose.

Cheers,

To construct α-helix you can try our tool

Conference Paper SeqOPT: web based server for rational design of conformation...

How can I access to your tools.

Thanks

Link was in abstract.

http://mml.spbstu.ru/seqopt