Local Similarity in RNA Secondary Structures

  • Authors:
  • Matthias Höchsmann;Thomas Töller;Robert Giegerich;Stefan Kurtz

  • Affiliations:
  • -;-;-;-

  • Venue:
  • CSB '03 Proceedings of the IEEE Computer Society Conference on Bioinformatics
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

We present a systematic treatment of alignment distanceand local similarity algorithms on trees and forests. Webuild upon the tree alignment algorithm for ordered treesgiven by Jiang et. al (1995) and extend it to calculate localforest alignments, which is essential for finding localsimilar regions in RNA secondary structures. The timecomplexity of our algorithm is O(|F1| \cdot |F2 \cdot deg(F1) \cdotdeg(F2) \cdot (deg(F1) + deg(F2)) where |Fi| is the number of nodes in forest Fi and deg (Fi) is the degree of Fi. Weprovide carefully engineered dynamic programming implementationsusing dense, two-dimensional tables which considerablyreduces the space requirement. We suggest anew representation of RNA secondary structures as foreststhat allow reasonable scoring of edit operations on RNAsecondary structures. The comparison of RNA secondarystructures is facilitated by a new visualization technique forRNA secondary structure alignments. Finally, we show howpotential regulatory motifs can be discovered solely by theirstructural preservation, and independent of their sequenceconservation and position.