Fully integrated SNV, fusion, and isoform analysis of cancer drivers with FLAIR3 for improved functional interpretation

Abstract

Background: Although the impact of genomic variants and changes in transcriptome regulation are often studied in isolation, their interaction is critical to understanding oncogenesis. We developed FLAIR3, which performs a comprehensive and integrated analysis of single nucleotide variants (SNVs), indels, gene fusions, and alternative splicing using long-read RNA sequencing (lrRNA-seq) and predicts their combined impact on amino acid sequence. Results: By applying FLAIR3 to lrRNA-seq datasets from osteosarcoma and lung adenocarcinoma, we demonstrated that long-read technology captures the full-length molecules required to phase variants with specific isoforms. In the osteosarcoma samples, FLAIR3 identified alternative splicing of gene fusions in known driver genes, including TP53. In the lung adenocarcinoma samples, FLAIR3 revealed isoform-biased expression of oncogenic SNVs, including BRAF, CDKN2A, and TP53. Conclusions: FLAIR3 provides a robust method for resolving the complex interplay between variants and transcript isoforms, offering deeper insights into the molecular drivers of cancer.