Detecting and characterizing fusions and tandem duplications in acute myeloid leukemia transcriptome assemblies using Barnacle

Chimeric transcripts are RNA molecules that cannot be explained by linear models of alternative splicing and can arise from events at either the DNA or the RNA level. Three types of chimeras, fusions, partial tandem duplications (PTDs), and internal tandem duplications (ITDs), are important in the detection, prognosis, and treatment of many human cancers. Here we report Barnacle, a high-throughput analysis tool that detects and characterizes fusions, PTDs, and ITDs in de novo assembled RNA-seq data. We characterized Barnacle’s sensitivity and specificity with simulated data, and compared Barnacle’s fusion detection performance with that of TopHat-Fusion. We ran Barnacle on two deeply-sequenced acute myeloid leukemia (AML) RNA-seq datasets. Among the events that Barnacle predicted in these libraries are three known to be important in AML: fusions between PML and RARA, PTDs in MLL, and ITDs in FLT3.