Fusions

Fusion Calling

DRAGEN fusion caller performs the fusion calling. The outputs are VCF files. It first scans the genome to discover evidence of possible structural variants (SV) and large indels based on split- and spanning-reads. The evidence is enumerated into a graph with edges that connect all regions of the genome that have a possible breakend association. It then analyzes individual graph edges to discover and score SVs associated with the edges. The substeps of this process include the following items:

• Inference of SV candidates associated with the edge.

• Attempted assembly of the SVs breakends.

• Scoring, genotyping, and filtering of the SV.

• Output to VCF.

Fusion Filtering

In assays, there are hundreds to thousands of fusion candidates in a single sample. Most fusion candidates (~99%) are false positives. The fusion filtering tool, DNA Fusion Filter (DNAFF), distinguishes true fusion calls from the false positives. DNAFF performs the following functions:

• Removes spurious fusions including fusions with only one supporting read pair and fusions that overlap with repeat regions, which are more likely to have sequencing errors.

• Filters nonconfident supporting reads for all fusion candidates based on the following criteria:

  • Filter reads with low-sequence identity with the fusion contig.

  • Filter erroneous reads, which are reads that do not support the fusion. For example, reads that have suspicious supplementary alignment.

  • Deduplicate reads based on UMI information.

• Applies the following rules for the final fusion output

  • If the fusion gene pair has been reported in COSMIC, it must have ≥ 2 supporting reads.

  • If the fusion gene pair has not been reported in COSMIC, it must have ≥ 3 supporting reads.

  • At least one fusion breakpoint must fall within the 23 target genes.

Fusion Output Files

Fusion calling results are output in several files:

1. Combined Variant Output File, {SampleID}_CombinedVariantOutput.tsv

2. Fusions CSV, {Sample_ID}_Fusions.csv

1. Combined Variant Output File

File name: {SampleID}_CombinedVariantOutput.tsv

Fusion calling results are output in the [DNA Fusions] section. Fusion variants must meet the following conditions to be included:

• Passing variant call (KeepFusion field is true).

• Contains at least one gene on the fusion allow list.

• Genes separated by a dash (-) indicate that the fusion directionality could be determined. Genes separated by a slash (/) indicate that the fusion directionality could not be determined.

2. Fusions CSV

File name: {Sample_ID}_Fusions.csv

The fusions file contains all candidate fusions identified by the analysis pipeline.

The fusion columns are described in the following table. If you use Microsoft Excel to view this file, genes that are convertible to dates (for example, MARCH1) automatically convert to dd‑mm format (1-Mar).

Fusion Object Field	Description
Sample	Input sample ID.
Name	Fusion name as reported by the DRAGEN fusion caller.
Chr1	The chromosome of the first breakend.
Pos1	The position of the first breakend.
Chr2	The chromosome of the second breakend.
Pos2	The position of the second breakend.
Direction	The direction of how the breakends are joined.
Alt_Depth	The number of read-pairs supporting the fusion call.
Total_Depth	Max number of read-pairs aligned to a fusion breakend.
BP1_Depth	Number of read-pairs aligned to the first breakend.
BP2_Depth	Number of read-pairs aligned to the second breakend.
VAF	Variant allele frequency.
Gene1	Genes that overlap the first breakend.
Gene2	Genes that overlap the second breakend.
Contig	The fusion contig.
Filter	Indicates whether the fusion has passed all of the fusion filters.
Is_Cosmic_GenePair	Indicates whether the gene pair has been reported by COSMIC (True/False).
Fusion Directionality Known	Indicates whether the fusion direction is known, and indicated by the order of the genes (True/False).

The following table lists the meaning of the values in the direction column. The values are in the format used by Samtools.

Direction	VCF Format	Description
L1R2	t[p[	The left of breakend1 is joined with the right of breakend2.
L1rL2	t]p]	The left of breakend1 is joined with the reverse complement of the left of breakend2.
L2R1	]p]t	The left of breakend2 is joined with the right of breakend1.
rR2R1	[p[t	The reverse complement of the right of breakend2 is joined with the right of breakend1.