Calculating correlated mutations at the protein interface: free energies of pairwise mutants

From Master Projects
Revision as of 09:09, 15 November 2013 by Feenstra (talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Calculating correlated mutations at the protein interface: free energies of pairwise mutants
status: finished
Master: project within::Bioinformatics
Student name: student name::Helen Kruize
Start start date:=2013/04/02
End end date:=2013/06/01
Supervisor: Anton Feenstra
Second supervisor: Qingzhen Hou
Second reader: has second reader::Qingzhen Hou
Thesis: has thesis::Media:Thesis.pdf
Poster: has poster::Media:Posternaam.pdf

Signature supervisor



Correlated mutations have been used for many years to identify functionally crucial residue-pairs, commonly interpreted as well as residue contacts. Recent developments in methods, but mainly the increased size of sequence databases, have made predictions of residue contacts accurate enough to be useful for protein structure and interface predictions.

In (other) recent work, we have showen that using molecular dynamics simulations and a coarse-grained forcefield, we can calculate the free energy barrier for unbinding with accuracy at least similar to that of (much more expensive) atomistic simulations.

This project will explore the feasibility of calculating residue correlations based on these free energy calculations. Starting with the MP1-p14 complex (PDB id 1vet), each 'worst case' mutation (according to the BLOSUM62 substitution matrix) will be examined. Mutations will be created using modeller. For each of these single point mutations a PMF will be calculated and the significant deviation of the PMF well depth will be determined. In a second stage, we will do the same for all interacting pairs of residues across the protein interface.

In a later stage, we can look at sequence alignmnents and the structure of the complex (for two proteins), we will identify interface contacts and corresponding pairs of mutations that do occurr in the sequence data, and other pairs of mutations that do not occur in the sequences. We will then establish if we, as we expect, indeed find a much lower impact on the free energy of the tolerated (occurring) pairwise mutations, and a much larger effect than the forbidden (not occurring) mutations.

Related projects

Modeling the Effective Interaction Between Proteins - A Coarse-Grained Approach (Ali May)