If you are using linux.

Then it takes a single command to convert pdb to fasta sequence.

Download the script: https://github.com/dongshuyan/pdb2fasta

Extract it inside a folder and place your pdb file inside it.

Rename your pdb file into sa.pdb

Open the folder with shell/command line interface.

And run the script:

python pdb2fasta.py sa.pdb > sa.fa

and it will produce sa.fasta automatically.

Rosetta Design http://rosettadesign.med.unc.edu (Article RosettaDesign server for protein design

William Ford

Check this servers:

https://swift.cmbi.umcn.nl/servers/html/soupir.html

http://www.meilerlab.org/index.php/servers/show?s_id=3

tool:

http://cupnet.net/pdb2fasta/

Hope this will help you.

Depending on what your need is - if you are working with a structure that has been deposited in the PDB, you may want to use the sequence associated with that file in the PDB, which (usually) is the sequence of the construct used - including residues that are not resolved in the structure because of disorder. This sequence is also to be found in the SEQRES records in the header of the PDB file. If you need exactly the sequence represented by the coordinates, the servers listed by

Sureshkumar v. will do a good job. Alternatively, it is usually possible to extract the sequence of a structure in many structure visualization programs, e.g. in PyMOL

you just specify "save name.fasta, object_or_selection" to create a fasta format sequence file of your object or selection. My first answer refers to the problem of designing de-novo proteins, where you create a desired backbone structure from secondary structure elements or fragments of real proteins, and then have to design a sequence that will stabilize the designed structure.

Just open that pdb file in PyMol and run this command: print cmd.get_fastastr('all')

That is called the "backwards folding problem" and it was shown to be solvable by Eisemberg's group (

Lüthy R, Bowie JU, Eisenberg D (1992) Assessment of protein models with three-dimensional profiles. Nature 356: 83–5.

https://www.nature.com/articles/356083a0

), and later a Rosetta module ("design" or "fixbb" depending on the version) was created to this aim (this was the starting motivation to program what today is know as ROSETTA suit) , giving one of the first spectacular hits of David Baker and his Group (

Kuhlman B, Dantas G, Ireton GC, Varani G, Stoddard BL, et al. (2003) Design of a novel globular protein fold with atomic-level accuracy. Science 302: 1364–68. http://science.sciencemag.org/content/302/5649/1364

).

You may find the following reference interesting about one possible application of the "backward folding problem":

Martínez-Castilla LP, Rodríguez-Sotres R (2010) A Score of the Ability of a Three-Dimensional Protein Model to Retrieve Its Own Sequence as a Quantitative Measure of Its Quality and Appropriateness. PLoS ONE 5(9): e12483.

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0012483

.

Best wishes,

Rogelio

There are many options and almost any software dealing with the representation of the coordinates will do it. You may try the freeware Chimera (https://www.cgl.ucsf.edu/chimera/) . it is simple and powerful.

I agree with Annemarie Honegger answer for this question

If you are using linux.

Then it takes a single command to convert pdb to fasta sequence.

Download the script: https://github.com/dongshuyan/pdb2fasta

Extract it inside a folder and place your pdb file inside it.

Rename your pdb file into sa.pdb

Open the folder with shell/command line interface.

And run the script:

python pdb2fasta.py sa.pdb > sa.fa

and it will produce sa.fasta automatically.

Dear SAMEE ULLAH ,

I do not think that was the question. I think the question was:

Given the three-dimensional coordinates of a protein backbone, Which amino acid sequences fit that backbone and maintain the protein fold stable?

William Ford please correct me if I am wrong.

Rogelio Rodríguez-Sotres Well as web server mentioned by Suresh kumar v. and pymol by Kazi i.e extract a seq from pdb structure. And Dr. Annemarie agree with both of them "If you need exactly the sequence represented by the coordinates(-Comment of Dr Annemarie)".

So the above script will do the same thing i.e extract a seq from pdb structre.

And i'm sure what @William Ford really mean of vice versa Forexample; In Swiss predict the structure from input sequence, while no option of out-put for the sequence of the predicted structure if some one really needs it.

Dear SAMEE ULLAH

As I commented in the previous contribution to this thread there are solutions to the so called "inverse folding problem (known structural fold to predicted sequence)", and these are more reliable (higher change of matching reality) than solutions to the "direct folding problem (known sequence to predicted structural fold)".

https://www.nature.com/articles/356083a0

http://science.sciencemag.org/content/302/5649/1364

https://journals.plos.org/plosone/article?id=10.1371/journal.pone.0012483

Currently the best algorithm is apparently the one implemented in the Rosetta suit as the "fixbb" protocol.

Best wishes

Rogelio