Computational pipeline for the empirical detection of cleaved targets of small silencing RNAs using degradome sequencing data

Open Access
Author:
Addo-Quaye, Charles
Graduate Program:
Computer Science and Engineering
Degree:
Doctor of Philosophy
Document Type:
Dissertation
Date of Defense:
July 14, 2009
Committee Members:
  • Webb Colby Miller, Dissertation Advisor
  • Webb Colby Miller, Committee Chair
  • Michael Axtell, Committee Chair
  • Ali Hurson, Committee Member
  • Raj Acharya, Committee Member
  • Padma Raghavan, Committee Member
  • John Edward Carlson, Committee Member
Keywords:
  • small RNAs microRNAs
Abstract:
Small silencing RNAs are short oligonucleotide sequences, usually 20-30 nucleotides long, and include microRNAs (miRNAs), short interfering RNAs (siRNAs), transacting siRNAs and Piwi-interacting RNAs (piRNAs). These small silencing RNAs play a critical role in the regulation of gene expression in the genomes of eukaryotes, by guiding gene silencing complexes to the appropriate target-gene transcripts. Confident identification of genes targeted by miRNAs is crucial to unraveling their biological relevance. In plants, miRNAs tend to have a high similarity to their target genes. Computer programs that search transcriptome-wide for sequence similarity to mature miRNA sequences have widely been used to predict miRNA target genes. Currently available programs have varying degrees of sensitivity and specificity and hence predicted target genes require experimental confirmation. Confirmation is time-consuming and can only be carried out on a one-at-a-time basis. In plants, the results of miRNA-guided, Argonaute-catalyzed cleavage of target genes are uncapped and polyadenylated fragments that are stable in vivo. Degradome sequencing is a transcriptome-wide method for capturing the 5' ends of these stable, diagnostic fragments of cleaved messenger RNA transcripts. Data from degradome sequencing provides empirical evidence for detecting cleaved miRNA targets and eliminates the requirement for experimental confirmation. I implemented an efficient and generalizable computational pipeline for transcriptome-wide empirical detection of cleaved targets of miRNAs. Using the computational pipeline, I was able to detect previously known targets of plant miRNAs and transacting siRNAs and also confirm previously predicted targets as well as detect novel targets.