Year: 2009
Type: dissertation
Abstract: Protein structure prediction is concerned with the prediction of a \nprotein's three dimensional structure from its amino acid sequence. \nSuch predictions are commonly performed by searching the possible \nstructures and evaluating each structure by using some scoring \nfunction. If it is assumed that the target protein structure \nresembles the structure of a known protein, the search space can be \nsignificantly reduced. Such an approach is referred to as \ncomparative structure prediction. When such an assumption is \nnot made, the approach is known as ab initio structure \nprediction. There are several difficulties in devising efficient \nsearches or in computing the scoring function. Many of these \nproblems have ready solutions from known mathematical methods. \nHowever, the problems that are yet unsolved have hindered structure \nprediction methods from more ideal predictions. \n \nThe objective of this study is to present a complete framework for \nab initio protein structure prediction. To achieve this, a new \nsearch strategy is proposed, and better techniques are devised for \ncomputing the known scoring functions. Some of the remaining \nproblems in protein structure prediction are revisited. Several of \nthem are shown to be intractable. In many of these cases, approximation \nmethods are suggested as alternative solutions. The primary issues addressed in this thesis \nare concerned with local structures prediction, structure assembly \nor sampling, side chain packing, model comparison, and structural \nalignment. For brevity, we do not elaborate on these problems here; \na concise introduction is given in the first section of this thesis. \n \nResults from these studies prompted the development of several \nprograms, forming a utility suite for ab initio protein \nstructure prediction. Due to the general usefulness of these \nprograms, some of them are released with open source licenses to \nbenefit the community.
Source: UWSpace (University of Waterloo)
Author: Shuai Cheng Li
Language: English
Subfield: Molecular Biology
Domain: Life Sciences
Open Access status: green
OpenAlex ID: W2581853149