Date of Award

Spring 2013

Document Type

Thesis

Degree Name

Master of Science in Bioinformatics - (M.S.)

Department

Computer Science

First Advisor

Jason T. L. Wang

Second Advisor

Zhi Wei

Third Advisor

Mei Liu

Fourth Advisor

Usman W. Roshan

Abstract

Non-coding RNAs (ncRNAs) are the functional RNA molecules that are involved in many biological processes including gene regulation, chromosome replication and RNA modification. Searching genomes using computational methods has become an important asset for prediction and annotation of ncRNAs. To annotate an individual genome for a specific family of ncRNAs, a computational tool is interpreted to scan through the genome and align its sequence segments to some structure model for the ncRNA family. With the recent advances in detecting an ncRNA in the genome, heuristic techniques are designed to perform an accurate search and sequence-structure alignment. This study uses a novel approach for such genome wide search of ncRNAs using the RNATOPS and Infernal software tools, which incorporates heuristic dynamic programming algorithms to carry out the sequence analysis using the profiles of RNA consensus secondary structures.

Genome wide search for ncRNAs from thirteen genomes is performed using RNATOPS and Infernal. The training set of ncRNA multiple sequence alignments is prepared from RFAM and homologous Genomes are retrieved from RNASTRAND database. Through the experiments, performance of each tool is analyzed and compared with respect to their ncRNA search accuracies. It is further interfered that Infernal, compared to RNATOPS, is more accurate in detecting an ncRNA in all the thirteen genomes tested.

Share

COinS