Name: Fabien P.E. Huard Course: PHD;STAT Department: Department of Statistics Staff Supervisor: Prof. Graham Wood Associate Supervisor: Andrzej Kozek Email Address: fhuard@efs.mq.edu.au Thesis Title Investigation of the sequential aspect in protein folding

Abstract

We propose here a sequential approach to the protein folding problem. Current de novo methods for protein fold prediction initiate their computations using a fully extended protein chain length, i.e. the full amino acid sequence. Some recent studies however have pointed out the sequential aspect in protein folding. Proteins are likely to fold co-translationally as the amino-acids chain elongates, reaching local temporary conformations. Furthermore, folding rates are much higher than translation rates.

We elongate iteratively the length of protein to be folded starting from the N-terminus. Amino acids are added one by one. Each time the chain length increases from k to (k + 1) amino acids, the conformation simulated is permitted to change. Therefore the (k + 1) fold must be an evolution of the previously predicted fold for k residues. New conformations are evaluated with the help of scoring functions, and can be then retained or not. The final fold of the protein is obtained once all residues are added.