Clustering Of Protein Fragments

Question

[Aga]

Hi Everyone,

I have a set of short protein fragments (4-10aa). I would like to cluster them by structural similarity but WITHOUT changing their original positions and in a sequence independent mode. Do you know any program that can do this?

Thanks, Aga

[Darked89]

Are these peptides part of larger proteins for which you got the structures & PDB files? Or you need to use something maybe like GROMACS to calculate structures first?

[Rm]

do they have structures? BTW are they have disulphide bonds in them?

[Aga]

Yes, they have structures. They are not part of the larger proteins and they do not have disulphide bonds.

Answer 1

[Rm]

See if "MaxCluster" of any use for your short fragments?

It is a tool for Protein Structure Comparison and Clustering.

[Aga]

Thanks, MaxCluster would be good but as far as I know it performs superposition of the initial structures. In my case I do not want to change initial positions of my peptides.

[Rm]

In super position "it wont change the structure" rather it will rearrange the structure relative to the "reference" coordinate position. But you will know how similar are they wrt each other.

[Aga]

I do not want to change their positions, I am interested in finding both similar positions and similar structures.

Actually I need a program that can calculate something like optimal RMSD but WITHOUT changing the original positions of the structures. All programs that I know do the superposition.

[Rm]

I think RMSD is not possible unless you superpose. RMSD will tell you how much you are deviating WRT reference. This is like how well your feet will fit into someone else's shoe.