Adapting And Enhancing Hybrid Computational Methods For RNA Secondary Structure Prediction
Loading...
Date
2011-12
Authors
al-Khatib, Ra'ed Mohammad Ali
Journal Title
Journal ISSN
Volume Title
Publisher
Universiti Sains Malaysia
Abstract
The secondary structure of RNA with pseudoknots is widely utilized for tracing the RNA
tertiary structure, which is a key to understanding the functions of the RNAs and their useful
roles in developing drugs for viral diseases. Experimental methods for determining RNA
tertiary structure are time consuming and tedious. Therefore, predictive computational approaches
are required. Predicting the most accurate and energy-stable pseudoknot RNA secondary
structure has been proven to be an NP-hard problem. This thesis presents a hybrid
method to predict the RNA pseudoknot secondary structures by combining detection methods
with dynamic programming algorithms. This hybrid method is further enhanced by adopting
the case-based reasoning (CBR) technique. Three different methods are proposed, (i) Bioinspired
swarm intelligence method (HPRna); (ii) Adaptive hybrid method (MSeeker); and (iii)
Fast parallel method (FGTSeeker), where each is an improvement to the previous method. The
proposed prediction methods were evaluated against other existing prediction methods using
the real native structures as the main factor of comparison. Results show that the three proposed
methods obtained more accurate pseudoknotted RNA secondary structures with better
performance, especially in predicting long sequences.
Description
Keywords
Enhancing hybrid computational methods , for RNA secondary structure prediction