Evolutionary optimisation for protein structure prediction using low resolution protein models

Project Title:

Evolutionary optimisation for protein structure prediction using low resolution protein models

Supervisor(s):

A/Prof Madhusudan Chetty and A/Prof Adil Baghirov

Contact person and email address:

A/Prof Madhusudan Chetty, madhu.chetty@federation.edu.au

A brief description of the project:

Evolutionary Algorithms (EAs) (e.g. as memetic algorithms, genetic algorithms etc.), are effectively used for solving optimisation problems in various domains. This encourages us to study their applications for solving problems in computational biology, e.g. the complex protein structure prediction (PSP) problem. The proposed ab initio methods predict protein folding without using structure information from any other protein for comparison. These approaches can be useful for an initial approximation of the protein structure and for the investigation of the dynamics that govern the protein folding process. Due to astronomically large and complex search space, exhaustive search for an optimal solution from amongst this large number of conformations is almost impossible. We propose to exploit the property of choosing different local searches at different evolutions making EA adaptive. The success of EAs (more specifically, Memetic Algorithm, MA) for PSP problem lies on the LS (Local Search), how successfully it can be implemented, and the successful utilisation of domain knowledge.