Paediatric B-ALL nanopore WTS, high-depth cohort, dominant-fusion classification (Lin et al. 2026 Table 3)

Stable ID: rna-fusion-20261009-protocol-lin2026-high-depth

areas

rna-transcriptomes

contexts

clinical_research

protocol

The dominant detected B-ALL fusion (most supporting reads) is compared with the sample's known fusion subtype. TP: concordant; TN: no fusion and none detected; FN: known fusion and none detected; FP: a B-ALL fusion detected that does not match or in a fusion-negative sample.

version

Table 3

metric

recall

limitations

FUSILLI was developed by the authors and is author_reported. Its filters (two supporting reads, gene distance and overlap) were set from analyses of candidate fusion-supporting reads (Methods 'FUSILLI'; Supplemental Figure S1), and the source does not say these were held out from the evaluated cohorts. FusionSeeker, JAFFAL and LongGF were run by the same authors, who did not develop them.; Per-sample classification by the dominant fusion among detected B-ALL list fusions (74 fusions); calls outside the list and secondary fusions are not scored, so this does not measure the review burden of a caller's full output.; A sample with a known fusion whose dominant call is a different fusion counts as a false positive, not a false negative, so sensitivity is not the fraction of known-fusion samples correctly typed.; Truth is the clinical genomic subtype from cytogenetics, FISH and short-read RNA-seq (CICERO, FusionCatcher and manual review).; Paediatric B-ALL only; PAX5::ZCCHC7 excluded from scoring; the authors note IGH::DUX4, KMT2A and ZNF384 rearrangements were not tested.; Caller versions are not printed; JAFFAL ran from the 'latest' container image.; 51 samples, 27 with a known B-ALL fusion; the samples are a subset of the low-depth cohort.

source locator

Table 3; Materials and Methods 'Fusion Performance Evaluations'

• uses data: Paediatric B-ALL nanopore PCR-cDNA whole-transcriptome sequencing (UNC, St Jude, ECOG-ACRIN)
• subject: computational_cost: long-read fusion callers (Lin et al. 2026)
• assessment: FUSILLI on paediatric B-ALL nanopore WTS, high-depth (Lin et al. 2026)
• assessment: FusionSeeker on paediatric B-ALL nanopore WTS, high-depth (Lin et al. 2026)
• assessment: JAFFAL on paediatric B-ALL nanopore WTS, high-depth (Lin et al. 2026)
• assessment: LongGF on paediatric B-ALL nanopore WTS, high-depth (Lin et al. 2026)
• assessed by: Select a tumour RNA fusion-detection workflow