Refer to RNA Analysis Methods for more information.
The splice variant VCF contains all candidate splice variants targeted by the analysis panel identified by the RNA analysis pipeline. You can apply the following filters for each variant call:
Refer to the headers in the output for more information about each column.
If available, each splice variant is annotated using the Illumina Annotation Engine. The following information is captured in the JSON:
HGNC Gene
Transcript
Exons
Introns
The all fusions CSV file contains all candidate fusions identified by the DRAGEN RNA pipeline. Two output columns in the file describe the candidate fusions: Filter and KeepFusion.
The following table describes the semicolon-separated output found in the Filter columns. The output is either a confidence filter or information only as indicated. If none of the confidence filters are triggered, the Filter column contains the output PASS, else it contains the output FAIL.
Filter Column Output
The KeepFusion column of the output has a value of TRUE when none of the confidence filters are triggered.
Refer to the headers in the output for more information about each column.
Fusion Columns
When using Microsoft Excel to view this report, genes that are convertible to dates (such as MARCH1 automatically convert to dd-mm format (1 Mar) by Excel. The following are fusion allow list genes:
ABL1
AKT3
ALK
AR
Canonical
Consequence
MIN_SUPPORT
Confidence filter
The fusion candidate has very few fusion supporting reads (< 5 supporting read pairs).
READ_THROUGH
Confidence filter
The breakpoints are cis neighbors (< 200 kbp) on the reference genome.
ANCHOR_SUPPORT
Information only
Read alignments of fusion supporting reads are not long enough (12 bp) at either of the two breakpoints.
HOMOLOGOUS
Information only
The candidate is likely a false candidate generated because the two genes involved have high gene homology.
LOW_ALT_TO_REF
Information only
The number of fusion supporting reads is < 1% of the number of reads supporting the reference transcript at either of the two breakpoints.
LOW_GENE_COVERAGE
Information only
Each breakpoint in an enriched gene has fewer than 125 bp with nonzero read coverage.
NO_COMPLETE_SPLIT_READS
Confidence filter
For every fusion-supporting split read, the total number of aligned bases across two breakpoints is less 60% of the read length.
UNENRICHED_GENE
Confidence filter
Neither of the two parent genes is in the enrichment panel.
Gene B Breakpoint
[Information only] The chromosome and offset of the Gene B side of the fusion.
Gene B Location
Location of the breakpoint within Gene B: - IntactExon—Matches exon boundary - BrokenExon—Inside an exon - Intronic—Within an intron - Intergenic—No gene overlap (currently excluded) If multiple genes in Gene B, then semicolon separated list of locations. This column is used internally to identify genes to report when a breakpoint occurs in a region overlapping multiple genes. Occasionally, additional values are listed for genes that were excluded from the GeneB list.
Gene B Sense
Boolean indicating whether left/right breakpoint order suggests fusion transcript is in the same sense of Gene B. If multiple genes are in Gene B, then semicolon separated list of bools.
Gene B Strand
Strand of Gene B, + for forward, - for reverse.
Score
The quality of fusion as determined by DRAGEN server.
Filter
The filter associated with the fusion as determined by the respective caller. Results from different callers are not equivalent.
Ref A Dedup
Gene A uniquely mapping reads paired across or split by the junction. Does not support fusion. Duplicate reads are not included.
Ref B Dedup
Gene B uniquely mapping reads paired across or split by the junction. Does not support fusion. Duplicate reads are not included.
Alt Split Dedup
Uniquely mapping reads split by the junction. Supports fusion. Duplicate reads are not included.
Alt Pair Dedup
Uniquely mapping reads paired across junction. Supports fusion. Duplicate reads are not included.
KeepFusion
The determination whether the fusion should be kept or dropped from the list of fusions.
Fusion Directionality Known
Whether fusion directionality is known and indicated by gene order.
AXL
BCL2
BRAF
BRCA1
BRCA2
CDK4
CSF1R
EGFR
EML4
ERBB2
ERG
ESR1
ETS1
ETV1
ETV4
ETV5
EWSR1
FGFR1
FGFR2
FGFR3
FGFR4
FLI1
FLT1
FLT3
JAK2
KDR
KIF5B
KIT
KMT2A
MET
MLLT3
MSH2
MYC
NOTCH1
NOTCH2
NOTCH3
NRG1
NTRK1
NTRK2
NTRK3
PAX3
PAX7
PDGFRA
PDGFRB
PIK3CA
PPARG
RAF1
RET
ROS1
RPS6KB1
TMPRSS2
LowQ
Splice variant score < passing quality score threshold value of 1.
PASS
Splice variant score ≥ passing quality score threshold value of 1.
LowUniqueAlignments
All splice junction supporting reads map to a unique genomic interval near at least one of the two splice sites.
DOUBLE_BROKEN_EXON
Confidence filter
If both breakpoints are distant from annotated exon boundaries, the number of supporting reads do not satisfy a high threshold requirement (≥ 10 supporting reads).
LOW_MAPQ
Confidence filter
All fusion supporting read alignments at either of the breakpoints have MAPQ < 20.
LOW_UNIQUE_ALIGNMENTS
Confidence filter
All fusion supporting read alignments map to a unique genomic interval at either of the breakpoints.
LOW_SCORE
Confidence filter
Gene A
The gene associated with the A side of the fusion. A semicolon delimited list is used for multiple genes.
Gene B
The gene associated with the B side of the fusion. A semicolon delimited list is used for multiple genes.
Gene A Breakpoint
[Information only] The chromosome and offset of the Gene A side of the fusion.
Gene A Location
Location of the breakpoint within Gene A: - IntactExon—Matches exon boundary - BrokenExon—Inside an exon - Intronic—Within an intron - Intergenic—No gene overlap (currently excluded) If multiple genes are in Gene A, then semicolon separated list of locations. This column is used internally to identify genes to report when a breakpoint occurs in a region overlapping multiple genes. Occasionally, additional values are listed for genes that were excluded from the GeneA list.
Gene A Sense
Boolean indicating whether left/right breakpoint order suggests fusion transcript is in the same sense of Gene A. If multiple genes are in Gene A, then semicolon separated list of bools.
Gene A Strand
Strand of Gene A, + for forward, - for reverse.
The fusion candidate has probabilistic score as determined by the features of the candidate.