Dear rajesh,

It is not possible to convert amino acid sequence directly into PDB file is in chimera. what is your intention behind the mentioning of basepair in amino acid sequence?

firstly to perform similarity search using BLASTP of your  AA sequence against PDB database(target database). If hit shows significant similarity(safe zone) with  query sequence then you can try homology modeling based server to obtain 3D structure (,pdb format) of your sequence  ie SWISSMODEL, Geno3D.

if target and template alignment shows insignificant  similarity score then u can try

Threading based servers for obtaining of 3D structure in PDB format ie phyre, SAM-T08 or Ab initio based methods ie HMMSTR/Rosetta ,

If the 3D structure of  target sequence is obtained by using any one above mentioned method/server, then compare two or more 3d structure by using DALI server, to find out structural variations.

Dear Rajesh, I do not understand your question. The amino acid sequence cannot be translated into a structure that easily. You can try Rosetta to obtain a plausible structure model of your sequence. However, from your question it appears that you are working with an enzyme that might have already been solved? in that case the structure should be deposited in the PDB database.

https://www.rosettacommons.org

http://www.rcsb.org

Dear rajesh,

It is not possible to convert amino acid sequence directly into PDB file is in chimera. what is your intention behind the mentioning of basepair in amino acid sequence?

firstly to perform similarity search using BLASTP of your  AA sequence against PDB database(target database). If hit shows significant similarity(safe zone) with  query sequence then you can try homology modeling based server to obtain 3D structure (,pdb format) of your sequence  ie SWISSMODEL, Geno3D.

if target and template alignment shows insignificant  similarity score then u can try

Threading based servers for obtaining of 3D structure in PDB format ie phyre, SAM-T08 or Ab initio based methods ie HMMSTR/Rosetta ,

If the 3D structure of  target sequence is obtained by using any one above mentioned method/server, then compare two or more 3d structure by using DALI server, to find out structural variations.

I agree with the above answers. As for my own answer, first I'd like to highlight Mahadev's question "What is the intention behind the mentioning of basepair? Are they inside the coding region? If yes then where are they located? At the terminal location? Sometimes the terminal chain of protein can not be determined due to higher movement (high beta value or high disorder) so they are missing. And if it is at the terminal position it is very less likely to contribute anything crucial to your enzyme.

To sum it up: the structure (of a protein) carries way more information than the sequence, so there's no simple "translation" available. Well, we know that the sequence determines the structure, but we don't know exactly HOW it's being determined. We can model what nature is doing to some degree, but it's still very far from perfect and will likely fail for longer structures/sequences. There's simply too many degrees of freedom to try and the fast methods we use to descrive the biological systems on moelcular level are too inaccurate. 

If you are lucky, a sequence that is similar to your sequence may have been structurally defined and published in the relevant databases.  Here is a link for a tutorial we built a few years ago that describes what homology modeling is and many of the available approaches... it assumes you have completed at least a biology undergraduate degree and that you have some understanding of what homology modeling is- its goals and limitations, etc from sitting in on the associated lecture or from some other source (¿start with wikipedia?)... i.e., tough to just hop right into this:

http://bioinformatics.burnham.org/SSBC/modeling.html

I would add that the newish Protein Model Portal provides a database of homology models created by other groups and it adds a system that can help you to assess the quality of the posted models or of new models that you produce through the PMP's web browser interface.  Begin by pasting your FASTA sequence into:

http://www.proteinmodelportal.org

For very short sequences, as detailed by others (Mahadev), you could try folding from scratch with Rosetta etc. ("short", I imagine, means less than 30-40 residues, unless you're an expert with Rosetta, but I haven't kept up with ab initio efforts).

Of course, beware, that all of these these tools will provide you with some "answer", but it is up to you to have learned how to assess whether than answer is reasonable or even plausible, so use them with caution!

 Exactly, how long is your protein sequence ? You mentioned 24-25 extra base pairs, I assume you mean extra amino acids. 

If your 25 residues are part of a larger protein, then try homology modelling.  This has been already suggested by others, but you could try Modeller for homology modelling. I would not look too much into ab-initio methods for prediction of large proteins.

If say you get a good quality model and still have some stretches of amino acids with no structural information (say about 10-20), then you could employ enhanced sampling methods like REMD and likes to obtain probable configurations for your protein. These methods are being extensively used for protein folding. But the caveat,  they require large computational efforts. Although, some methods (see Replica exchange with Solute Scaling) allow you to apply the sampling methods only to a part of your protein which may help in reducing the amount of calculations.

Some useful links:

http://www.ncbi.nlm.nih.gov/pubmed/21714551

http://www.gromacs.org/Documentation/How-tos/REMD

https://salilab.org/modeller/

Simply you can use Ribosome to convert your raw amino acid sequence to .pdb format. See the examples present in downloaded Ribosome software. Ribosome will work on Linux system.

all d best. 

you can convert it with a prediction approach in this site:

http://pasilla.health.unm.edu/tomcat/biocomp/convert

Or by this server that is better:

http://mobyle.rpbs.univ-paris-diderot.fr/cgi-bin/portal.py#welcome

Hi, If you plan to predict a protein/peptide structure from an amino acid sequence using an in silico approach, you can try using I-TASSER: https://zhanglab.ccmb.med.umich.edu/I-TASSER/.

You would be able to then generate a predictive structure and PDB to view in Pymol.

Monray