There are a variety of programs available for computationally introducing

mutations into protein structures and analyzing their potential effects on

structure and stability. They differ in their level of sophistication:

You might introduce a mutant using PyMOL, a program mainly geared at

visualization of macromolecular structures, use any of the homology modelling

programs e.g. MODELLER, as stand-alone command line application or embedded in

a viewing application such as chimera, or introduce and analyze mutations in

more complex modelling and analysis packages, such as the Rosetta suite

https://www.rosettacommons.org

and a variety of other software packages, ranging from those that are free for

academic use to quite costly and are targeted to industrial customers.

Which ones are best for you depends on your level of expertise and on the

complexity of your the tasks.

(See: https://en.wikipedia.org/wiki/Category:Molecular_modelling_software)

PDB files are ASCII text files,organized in fixed length, fixed format records

of 80 characters in length: the first 6 characters determine the type of record.

Open a pdb file with a plain text editor (e.g Notepad++ under Win, TextWrangler

or BBedit on Mac). You'll find first the header containing general information

on the molecule and experimental detail, followed by the coordinate DATA. For a

comprehensive description of the PDB format, see:

http://www.wwpdb.org/documentation/file-format

The pdb format, based on the 80 character limit of punched cards, is being

phased put in favor of the PDBx/mmCIF format, an XML format based on the PDB

markup language described here:

http://mmcif.wwpdb.org

they also provide links to a variety of software ressources:

http://mmcif.wwpdb.org/docs/software-resources.html

In particular, BioPython, a software toolkit for computational biology, provides

a library of tools for manipulating pdb and mmCIF files for python programmers,

while BioJava does the same for Java.

Hi Ryosaku,

I am using Coot (https://www2.mrc-lmb.cam.ac.uk/personal/pemsley/coot/) for modeling. It should be free for academics. It has all what you need. You can visualize, mutate, choose the side-chain conformation.

It has an option to fetch the structure directly from Protein Data Bank and you can safe those coordinates locally. You can superpose several structures using different algorithms, and make some figures (not as fancy as by PyMOL).

I am attaching one of the figures made in Coot, and used in publication. There are electron density maps and part of the protein chain.

George.

Hi Ryosaku,

I would like to add a warning to the useful answers already given. Indeed, many software will let you mutate a residue into another... The resulting model will, in many cases, be a good approximation of reality. However, you should keep in mind that mutating a residue in a protein can lead to some unpredictable results... it can destabilize the protein, locally change its structure, and so on. Should you want to derive useful information from your model, you should also consider an experimental confirmation of the structural integrity of the protein (using circular dichroism, NMR, thermal shift analysis...).

Laurent

you can do, but you have to see about the substituted amino acids i,e whether they are important for the stability of the structure or not. Further computationally u can validate by MD simulation.