Nextflow in Bioinformatics: Executors Performance Comparison Using Genomics Data

Spišaková, Viktória; Hejtmánek, Lukáš; Hynšt, Jakub

doi:10.1016/j.future.2023.01.009

Cited by 3 publications

(1 citation statement)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Current mRNAseq analyses are time-consuming, at times exceeding 20 h to analyse one sample [25], and require significant personnel, specialised expertise, and computational resources to establish and execute them in a routine clinical context. In recent years, the Bioinformatics and Clinical Genomics community has attempted to address these issues by using state-of-theart analysis workflow tools that can be implemented in routine sequencing and research labs [26,27]. However the careful consideration of parameter and filtering selection may still be required to achieve detection sensitivity to potential gene fusion events.…”

Section: Methodsmentioning

confidence: 99%

Reproducible Bioinformatics Analysis Workflows for Detecting IGH Gene Fusions in B-Cell Acute Lymphoblastic Leukaemia Patients

Thomson,

Rehn,

Heatley

et al. 2023

Cancers

View full text Add to dashboard Cite

B-cell acute lymphoblastic leukaemia (B-ALL) is characterised by diverse genomic alterations, the most frequent being gene fusions detected via transcriptomic analysis (mRNA-seq). Due to its hypervariable nature, gene fusions involving the Immunoglobulin Heavy Chain (IGH) locus can be difficult to detect with standard gene fusion calling algorithms and significant computational resources and analysis times are required. We aimed to optimize a gene fusion calling workflow to achieve best-case sensitivity for IGH gene fusion detection. Using Nextflow, we developed a simplified workflow containing the algorithms FusionCatcher, Arriba, and STAR-Fusion. We analysed samples from 35 patients harbouring IGH fusions (IGH::CRLF2 n = 17, IGH::DUX4 n = 15, IGH::EPOR n = 3) and assessed the detection rates for each caller, before optimizing the parameters to enhance sensitivity for IGH fusions. Initial results showed that FusionCatcher and Arriba outperformed STAR-Fusion (85–89% vs. 29% of IGH fusions reported). We found that extensive filtering in STAR-Fusion hindered IGH reporting. By adjusting specific filtering steps (e.g., read support, fusion fragments per million total reads), we achieved a 94% reporting rate for IGH fusions with STAR-Fusion. This analysis highlights the importance of filtering optimization for IGH gene fusion events, offering alternative workflows for difficult-to-detect high-risk B-ALL subtypes.

show abstract

Section: Methodsmentioning

confidence: 99%