Prediction of − Turns Using Global Adaptive Techniques from Multiple Alignments in Neural Network

Zainuddin, Zarita and Chan, Siow Cheng and Lye, Weng Kit (2008) Prediction of − Turns Using Global Adaptive Techniques from Multiple Alignments in Neural Network. Malaysian Journal of Mathematical Sciences, 2 (2). pp. 185-194. ISSN 1823-8343

[img] PDF
95Kb

Abstract

A neural network-based method has been developed for the prediction of − turns in proteins by using multiple sequence alignment. A feed-forward network with a single hidden layer is used where the sequence structure network is trained with multiple sequence alignment in the form of position-specific scoring matrices (PSSM). This paper concentrates on global adaptive techniques: Conjugate Gradient methods (Fletcher Reeves, Polak Ribiere and Powell Beale), Preconditioned Conjugate Gradient methods (Preconditioned Fletcher Reeves, Preconditioned Polak Ribiere and Preconditioned Powell Beale) and Levenberg Marquardt method in the training of multilayer perceptrons (MLP) neural network. The behavior of these training methods in the present study is reported. The Levenberg Marquardt method had been proved to be the most effective when tested and compared to other methods. It only takes 11 iterations to converge and yields overall performance, Qtotal = 96.67%. Then Preconditioned Fletcher Reeves, Preconditioned Polak Ribiere and Preconditioned Powell Beale methods yield the same result of total Q which is 93.33% with 333, 404 and 354 epochs accordingly. Lastly, Fletcher Reeves method yields the overall prediction accuracy of 93.33% after 689 epochs. On the other hand, Polak Ribiere and Powell Beale methods yield the same result of total Q which is 90% with 531 and 468 iterations respectively.

Item Type:Article
Keyword:neural network, prediction, secondary structure, multiple alignment protein
Faculty or Institute:Institute for Mathematical Research
Publisher:UPM Press
ID Code:12610
Deposited By: kmportal
Deposited On:09 Jun 2011 09:50
Last Modified:27 May 2013 07:53

Repository Staff Only: Edit item detail

Document Download Statistics

This item has been downloaded for since 09 Jun 2011 09:50.

View statistics for "Prediction of − Turns Using Global Adaptive Techniques from Multiple Alignments in Neural Network"


Universiti Putra Malaysia Institutional Repository

Universiti Putra Malaysia Institutional Repository is an on-line digital archive that serves as a central collection and storage of scientific information and research at the Universiti Putra Malaysia.

Currently, the collections deposited in the IR consists of Master and PhD theses, Master and PhD Project Report, Journal Articles, Journal Bulletins, Conference Papers, UPM News, Newspaper Cuttings, Patents and Inaugural Lectures.

As the policy of the university does not permit users to view thesis in full text, access is only given to the first 24 pages only.