deSALT: fast and accurate long transcriptomic read alignment with de Bruijn graph-based index

Long-read RNA sequencing (RNA-Seq) is promising to transcriptomics studies, however, the alignment of the reads is still a fundamental but non-trivial task due to the sequencing errors and complicated gene structures. We propose deSALT, a tailored two-pass long RNA-seq read alignment approach, which constructs graph-based alignment skeletons to sensitively infer exons, and use them to generate spliced reference sequence to produce refined alignments. deSALT addresses several difficult issues, such as small exons, serious sequencing errors and consensus spliced alignment. Benchmarks demonstrate that this approach has a better ability to produce high-quality full-length alignments, which has enormous potentials to transcriptomics studies.

The source code of deSALT and the data simulation and benchmarking scripts are available at: https://github.com/hitbc/deSALT.