STARFusion

STARFusion is an open-source software tool designed for the assembly of long-read sequencing data, particularly useful in genomic and metagenomic research. Developed by the University of California, Santa Cruz (UCSC) Genome Bioinformatics Group, STARFusion combines the strengths of multiple assembly algorithms to improve the accuracy and reliability of fusion gene detection. Fusion genes are abnormal junctions formed by the joining of two separate genes, often associated with cancer and other diseases.

The software integrates several key features to enhance its performance. It employs a two-step process: first, it aligns sequencing reads to a reference genome using a high-performance aligner like STAR (Spliced Transcripts Alignment to a Reference). Second, it identifies potential fusion events by analyzing discordant read pairs and split reads. STARFusion then refines these candidates through a graph-based assembly approach, which helps resolve complex fusion structures that might be missed by simpler methods.

STARFusion is particularly effective for analyzing data from platforms like PacBio and Oxford Nanopore Technologies, which generate long reads capable of spanning entire exons or even entire genes. This capability is critical for detecting fusions that involve large genomic rearrangements or complex splice junctions. The tool also provides detailed output, including graphical representations of fusion events and supporting evidence from sequencing reads, aiding researchers in validating and interpreting their findings.

STARFusion is freely available under an open-source license, making it accessible to both academic and clinical research communities. It is widely used in cancer genomics, where fusion genes are known to play a significant role in tumorigenesis. The software’s integration with common bioinformatics workflows and its compatibility with various sequencing technologies have contributed to its adoption as a standard tool in genomic research.