Identifying similar transcripts in a related organism from de Bruijn graphs of RNA-Seq data, with applications to the study of salt and waterlogging tolerance in melilotus

Abstract

While a popular strategy in de novo transcriptome assembly algorithms is to assemble the reads by obtaining a de Bruijn graph that represents the transcriptome, an additional step is needed to obtain predicted transcripts from the de Bruijn graph. A similarity search algorithm is then applied to a related organism to obtain information about possible function of these predicted transcripts. We observe that it is possible to obtain a more complete set of similar transcripts by starting the search from the de Bruijn graph directly. We develop a heuristic extension algorithm to identify paths in the de Bruijn graph that are similar to transcripts from the related organism. The algorithm starts by enumerating short paths in the de Bruijn graph, and makes use of evolutionary information from the related organism to iteratively extend these paths in the most promising direction. By extracting reads from publicly available RNA-Seq libraries, we apply our algorithm on both model organisms and non-model organisms to identify similar transcripts to related organisms with varying evolutionary distances. We show that our algorithm is able to recover more similar transcripts than existing algorithms, with longer transcripts and a better resolution of isoforms. We construct new RNA-Seq libraries for two Melilotas species, and apply our algorithm to study salt and waterlogging tolerance in these two species.