Chaining Sequence/Structure Seeds for Computing RNA Similarity

Laetitia Bourgeade (1) , Cedric Chauve (2) , Julien Allali (1)

1 LaBRI - Laboratoire Bordelais de Recherche en Informatique
2 SFU - Department of Mathematics [Burnaby]

We describe a new method to compare a query RNA with a static set of target RNAs. Our method is based on (i) a static indexing of the sequence/structure seeds of the target RNAs, (ii) searching the target RNAs by detecting seeds of the query present in the target, chaining these seeds in promising candidate homologs, then (iii) completing the alignment using an anchor-based exact alignment algorithm. We apply our method on the benchmark Bralibase2.1 and compare its accuracy and efficiency with the exact method LocaRNA and its recent seeds-based speed-up ExpLoc-P. Our pipeline RNA-unchained greatly improves computation time of LocaRNA and is comparable to the one of ExpLoc-P, while improving the overall accuracy of the final alignments.

Domaines: Bio-informatique [q-bio.QM]; Bio-Informatique, Biologie Systémique [q-bio.QM]; Algorithme et structure de données [cs.DS]

Laetitia Bourgeade, Cedric Chauve, Julien Allali. Chaining Sequence/Structure Seeds for Computing RNA Similarity. First Workshop on Computational Methods for Structural RNAs - CMRS'14, Sep 2014, Strasbourg, France. ⟨hal-01019797⟩