Secondary structure predictor

• Main Author: Kong, Benjamin Xupeng
• Other Authors: Kwoh Chee Keong
• Format: Final Year Project
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Computer science and engineering::Computer applications::Computers in other systems
• Online Access: http://hdl.handle.net/10356/62710
• Institution: Nanyang Technological University

In bioinformatics, secondary structure predictors are tools that are used to determine the secondary structure conformations of proteins based only on their amino acid sequence (primary structure). This means predicting the likelihood of sequences that will form secondary structures such as alpha helices, beta strands and coiled coils. There are two different methods in secondary structure prediction. The first method is the sequenced-based method. This method only requires the amino acid sequence of a protein. Inputs can be in string representative format, e.g. - FASTA or PDB format. The second method is the structure-based method. This method traces the backbone of a given protein structure to identify regions that will form solid secondary structure elements. It requires the specific positions of the atoms of the different amino acid residues. The first part of this project is to research on the different predictor tools that implement sequence-based or structure-based method. The second part of the project is to characterize weak points in the secondary structures that will break under stress or when given higher kinetic energy. This project aims to be a corner stone for logically dividing a macromolecule into smaller pieces of secondary structures which can be then used as coarse grains for molecular dynamics (MD) simulation. It is necessary to break a macromolecule into smaller pieces as MD simulations on very large molecules may require large computer resources and maybe computationally expensive. Furthermore, not every part of a macromolecule is needed to be observed during MD simulation. It is better to use reduced representations (coarse-grained models). Hence, this project implements a wrapper that is able to carry out sequence-based or structure-based prediction on inputs to identify the important regions of a protein secondary structure thus dividing them into smaller pieces.